Bioconjugation Protocols: Strategies and Methods, 2nd Edition (Methods in Molecular Biology Vol 751)

Methods

in

Molecular Biology™

Series Editor John M. Walker School of Life Sciences University of Hertfordshire Hatfield, Hertfordshire, AL10 9AB, UK



For further volumes: http://www.springer.com/series/7651

Bioconjugation Protocols Strategies and Methods

Second Edition

Edited by

Sonny S. Mark Pacific Biosciences, Menlo Park, CA, USA

Editor Sonny S. Mark, Ph.D. Pacific Biosciences Menlo Park CA, USA [email protected]

ISSN 1064-3745 e-ISSN 1940-6029 ISBN 978-1-61779-150-5 e-ISBN 978-1-61779-151-2 DOI 10.1007/978-1-61779-151-2 Springer New York Dordrecht Heidelberg London Library of Congress Control Number: 2011928240 © Springer Science+Business Media, LLC 2011 All rights reserved. This work may not be translated or copied in whole or in part without the written permission of the publisher (Humana Press, c/o Springer Science+Business Media, LLC, 233 Spring Street, New York, NY 10013, USA), except for brief excerpts in connection with reviews or scholarly analysis. Use in connection with any form of nformation storage and retrieval, electronic adaptation, computer software, or by similar or ­dissimilar methodology now known or hereafter developed is forbidden. The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights. While the advice and information in this book are believed to be true and accurate at the date of going to press, ­neither the authors nor the editors nor the publisher can accept any legal responsibility for any errors or omissions that may be made. The publisher makes no warranty, express or implied, with respect to the material contained herein. Printed on acid-free paper Humana Press is part of Springer Science+Business Media (www.springer.com)

Preface I am very pleased to present the second edition of Bioconjugation Protocols: Strategies and Methods, part of the excellent Methods in Molecular Biology™ book series. This current volume builds on the outstanding first edition originally conceived and developed by Prof. Christof M. Niemeyer at Technische Universität Dortmund (Germany). The first edition of this book aimed to address the deficiencies of many of the conventional approaches to the synthesis of chemically modified biomolecular conjugates that lack efficient means to control the stoichiometry of conjugation, as well as the specific site of attachment of the conjugated moiety. In keeping with that aim, this updated and expanded second edition of Bioconjugation Protocols further explores newer approaches that overcome the limitations of classical synthetic methods. In addition, a number of protocols collected in this new volume clearly reflect how insightful techniques and innovative approaches in bioconjugate chemistry can be derived from the seamless interplay between the fields of organic synthesis, surface biotechnology, nanobioscience, and materials science and engineering. It is thus my sincere hope that this revised edition of Bioconjugation Protocols continues to serve as a highly useful and practical reference for scientists of all disciplines confronting the challenges of semisynthesizing novel types of biomolecular reagents and/or biofunctionalizing surfaces and structures of unique interest for a variety of applications ranging from biomedical diagnostics to therapeutics and to biomaterials. The book is divided into five main parts, with Chaps. 1–24 in Parts I, II, and III describing the most recent, leading-edge approaches developed by researchers to prepare semisynthetic conjugates of native/modified biomacromolecules (proteins, nucleic acids, lipids, and carbohydrates). In Part IV, Chaps. 25–31 present methods for the preparation of biofunctionalized inorganic surfaces and polymer thin-film structures. And finally, Chaps. 32–36 in the last part of this book (Part V) specifically focus on procedures for the biofunctionalization of various types of metallic/semiconductor nanoparticles and other nanostructures (magnetic nanoparticles, quantum dots, carbon nanotubes, and silicon nanowire devices). I am most grateful to our distinguished group of international scholars who have generously and enthusiastically contributed their valuable time, tireless efforts, and expertise to make this reference volume truly unique and relevant. I would also like to express my thanks to Prof. John Walker, the Methods in Molecular Biology™ Series Editor at Springer Publishing, for his excellent editorial guidance during all stages of this book project. And finally, I sincerely thank my dear family and friends for their boundless patience and understanding and for kindly providing me with their support and encouragement during the production of this work. Menlo Park, CA

Sonny S. Mark

v

Contents Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

v xi

Part I  Protein Conjugates   1  Site-Specific Labeling of Proteins for Single-Molecule FRET Measurements Using Genetically Encoded Ketone Functionalities . . . . . . . . . . . . . . . . . . . . . . . . 3 Edward A. Lemke   2  Enzymatically Catalyzed Conjugation of a Biodegradable Polymer to Proteins and Small Molecules Using Microbial Transglutaminase . . . . . . . . . . . 17 Ahmed Besheer, Thomas C. Hertel, Jörg Kressler, Karsten Mäder, and Markus Pietzsch   3  Synthesis of Drug/Dye-Incorporated Polymer–Protein Hybrids . . . . . . . . . . . . . 29 Sukanta Dolai, Wei Shi, Bikash Mondal, and Krishnaswami Raja   4  Dye/DNA Conjugates as Multiple Labels for Antibodies in Sensitive Fluorescence Immunoassays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 Qin Zhang, Shengchao Zhu, and Liang-Hong Guo   5  Chemoselective Modification of Viral Proteins Bearing Metabolically Introduced “Clickable” Amino Acids and Sugars . . . . . . . . . . . . . . . . . . . . . . . . . 55 Partha S. Banerjee and Isaac S. Carrico   6  Preparation of Peptide and Other Biomolecular Conjugates Through Chemoselective Ligations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 Mathieu Galibert, Olivier Renaudet, Didier Boturyn, and Pascal Dumy   7  New Fluorescent Substrates of Microbial Transglutaminase and Its Application to Peptide Tag-Directed Covalent Protein Labeling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 Noriho Kamiya and Hiroki Abe   8  Covalent Conjugation of Poly(Ethylene Glycol) to Proteins and Peptides: Strategies and Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 Anna Mero, Chiara Clementi, Francesco M. Veronese, and Gianfranco Pasut   9  Extending the Scope of Site-Specific Cysteine Bioconjugation by Appending a Prelabeled Cysteine Tag to Proteins Using Protein Trans-Splicing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131 Tulika Dhar, Thomas Kurpiers, and Henning D. Mootz

Part II Nucleic Acid Conjugates 10  Polyethylenimine Bioconjugates for Imaging and DNA Delivery In Vivo . . . . . . . 145 Andrea Masotti and Francesco Pampaloni

vii

viii

Contents

11  Synthesis of a Glycomimetic Oligonucleotide Conjugate by 1,3-Dipolar Cycloaddition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Gwladys Pourceau, Albert Meyer, Jean-Jacques Vasseur, and François Morvan 12  Site-Specific DNA Labeling by Staudinger Ligation . . . . . . . . . . . . . . . . . . . . . . . Samuel H. Weisbrod, Anna Baccaro, and Andreas Marx 13  Improved Cellular Uptake of Antisense Peptide Nucleic Acids by Conjugation to a Cell-Penetrating Peptide and a Lipid Domain . . . . . . . . . . . . Takehiko Shiraishi and Peter E. Nielsen 14  Synthesis of Oligonucleotide–Peptide Conjugates for Biomedical and Technological Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Anna Aviñó, Santiago Grijalvo, Sónia Pérez-Rentero, Alejandra Garibotti, Montserrat Terrazas, and Ramon Eritja 15  Amphiphilic DNA Block Copolymers: Nucleic Acid-Polymer Hybrid Materials for Diagnostics and Biomedicine . . . . . . . . . . . . . . . . . . . . . . . . Jan Zimmermann, Minseok Kwak, Andrew J. Musser, and Andreas Herrmann

167

195

209

223

239

Part III Glycosyl and Lipid Conjugates 16  Chemically Selective Liposome Surface Glyco-functionalization . . . . . . . . . . . . . . Hailong Zhang, Yong Ma, and Xue-Long Sun 17  Bioconjugation Using Mutant Glycosyltransferases for the Site-Specific Labeling of Biomolecules with Sugars Carrying Chemical Handles . . . . . . . . . . . . Boopathy Ramakrishnan, Elizabeth Boeggeman, Marta Pasek, and Pradman K. Qasba 18  Lipid-Core-Peptide System for Self-Adjuvanting Synthetic Vaccine Delivery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Mariusz Skwarczynski and Istvan Toth 19  Coupling Carbohydrates to Proteins for Glycoconjugate Vaccine Development Using a Pentenoyl Group as a Convenient Linker . . . . . . . . . . . . . . Qianli Wang and Zhongwu Guo 20  Conjugation of LPS-Derived Oligosaccharides to Proteins Using Oxime Chemistry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Joanna Kubler-Kielb 21  Site-Specific Chemical Modification of a Glycoprotein Fragment Expressed in Yeast . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Junpeng Xiao and Thomas J. Tolbert 22  On-Resin Convergent Synthesis of a Glycopeptide from HIV gp120 Containing a High Mannose Type N-Linked Oligosaccharide . . . . . . . . . . . . . . . Rui Chen and Thomas J. Tolbert 23  Design and Synthesis of Novel Functional Lipid-Based Bioconjugates for Drug Delivery and Other Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Rupa R. Sawant and Vladimir P. Torchilin

269

281

297

309

317

329

343

357

Contents

ix

Part IV  Biofunctionalization of Surfaces and Thin Films 24  Chemical Functionalization and Bioconjugation Strategies for Atomic Force Microscope Cantilevers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Magnus Bergkvist and Nathaniel C. Cady 25  Chemoselective Protein and Peptide Immobilization on Biosensor Surfaces . . . . . Edith H.M. Lempens, Brett A. Helms, and Maarten Merkx 26  Fabrication of Dynamic Self-Assembled Monolayers for Cell Migration and Adhesion Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Nathan P. Westcott and Muhammad N. Yousaf 27  DNA Detection Using Functionalized Conducting Polymers . . . . . . . . . . . . . . . . Jadranka Travas-Sejdic, Hui Peng, Hsiao-hua Yu, and Shyh-Chyang Luo 28  Preparation and Dynamic Patterning of Supported Lipid Membranes Mimicking Cell Membranes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Stefan Kaufmann, Karthik Kumar, and Erik Reimhult 29  Enzyme Immobilization on Reactive Polymer Films . . . . . . . . . . . . . . . . . . . . . . Ana L. Cordeiro, Tilo Pompe, Katrin Salchert, and Carsten Werner 30  Characterization of Protein–Membrane Binding Interactions via a Microplate Assay Employing Whole Liposome Immobilization . . . . . . . . . . . . . Matthew D. Smith and Michael D. Best 31  A Bioconjugated Phospholipid Polymer Biointerface with Nanometer-Scaled Structure for Highly Sensitive Immunoassays . . . . . . . . . . . . . . . . . . . . . . . . . . . . Kazuki Nishizawa, Madoka Takai, and Kazuhiko Ishihara

381 401

421 437

453 465

477

491

Part V  Biofunctionalization of Nanostructures 32  Purification, Functionalization, and Bioconjugation of Carbon Nanotubes . . . . . John H.T. Luong, Keith B. Male, Khaled A. Mahmoud, and Fwu-Shan Sheu 33  Functional Integration of Membrane Proteins with Nanotube and Nanowire Transistor Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Aleksandr Noy, Alexander B. Artyukhin, Shih-Chieh Huang, Julio A. Martinez, and Nipun Misra 34  Single-Step Conjugation of Antibodies to Quantum Dots for Labeling Cell Surface Receptors in Mammalian Cells . . . . . . . . . . . . . . . . . . . . . . Gopal Iyer, Jianmin Xu, and Shimon Weiss 35  A Practical Strategy for Constructing Nanodrugs Using Carbon Nanotubes as Carriers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Wei Wu and Xiqun Jiang 36  Design and Synthesis of Biofunctionalized Metallic/Magnetic Nanomaterials . . . Eun-Kyung Lim, Seungjoo Haam, Kwangyeol Lee, and Yong-Min Huh

505

533

553

565 583

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 597

Contributors Hiroki Abe  •  Department of Applied Chemistry, Graduate School of Engineering, Kyushu University, Fukuoka, Japan Alexander B. Artyukhin  •  School of Natural Sciences, University of California, Merced, CA, USA Anna Aviñó  •  Institute for Research in Biomedicine, IQAC-CSIC, CIBER-BBN Networking Centre on Bioengineering, Biomaterials and Nanomedicine, Barcelona, Spain Anna Baccaro  •  Department of Chemistry, Konstanz Research School Chemical Biology, University of Konstanz, Konstanz, Germany Partha S. Banerjee  •  Department of Chemistry, State University of New York, Stony Brook, NY, USA Magnus Bergkvist  •  College of Nanoscale Science and Engineering, University at Albany, Albany, NY, USA Ahmed Besheer  •  Department of Pharmacy, Ludwig-Maximilians-University Munich, Munich, Germany Michael D. Best  •  Department of Chemistry, The University of Tennessee, Knoxville, TN, USA Elizabeth Boeggeman  •  Structural Glycobiology Section, Center for Cancer Research Nanobiology Program, National Cancer Institute, Frederick, MD, USA; Basic Science Program, SAIC-Frederick, Inc., Frederick, MD, USA Didier Boturyn  •  Département de Chimie Moléculaire, Université de Grenoble, Grenoble, France Nathaniel C. Cady  •  College of Nanoscale Science and Engineering, University at Albany, Albany, NY, USA Isaac S. Carrico  •  Department of Chemistry, State University of New York, Stony Brook, NY, USA Rui Chen  •  Department of Chemistry, Indiana University, Bloomington, IN, USA Chiara Clementi  •  Department of Pharmaceutical Sciences, University of Padova, Padova, Italy Ana L. Cordeiro  •  Leibniz Institute of Polymer Research Dresden, Max Bergmann Center of Biomaterials Dresden, Dresden, Germany Tulika Dhar  •  Fakultät Chemie – Chemische Biologie, Technische Universität Dortmund, Dresden, Germany Sukanta Dolai  •  Department of Chemistry, The City University of New York at the College of Staten Island, Staten Island, NY, USA Pascal Dumy  •  Département de Chimie Moléculaire, Université de Grenoble, Grenoble, France

xi

xii

Contributors

Ramon Eritja  •  Institute for Research in Biomedicine, IQAC-CSIC, CIBER-BBN Networking Centre on Bioengineering, Biomaterials and Nanomedicine, Barcelona, Spain Mathieu Galibert  •  Département de Chimie Moléculaire, Université de Grenoble, Grenoble, France Alejandra Garibotti  •  Institute for Research in Biomedicine, IQAC-CSIC, CIBER-BBN Networking Centre on Bioengineering, Biomaterials and Nanomedicine, Barcelona, Spain Santiago Grijalvo  •  Institute for Research in Biomedicine, IQAC-CSIC, CIBER-BBN Networking Centre on Bioengineering, Biomaterials and Nanomedicine, Barcelona, Spain Zhongwu Guo  •  Department of Chemistry, Wayne State University, Detroit, MI, USA Liang-Hong Guo  •  Research Center for Eco-environmental Sciences, Chinese Academy of Sciences, Beijing, P. R. China Seungjoo Haam  •  Department of Chemical and Biomolecular Engineering, Yonsei University, Seoul, Korea Brett A. Helms  •  Laboratory of Chemical Biology, Department of Biomedical Engineering, Eindhoven University of Technology, Eindhoven, The Netherlands Andreas Herrmann  •  Department of Polymer Chemistry, The Zernike Institute for Advanced Materials, University of Groningen, Groningen, The Netherlands Thomas C. Hertel  •  Institute of Pharmacy, Martin Luther University, Halle Wittenberg, Germany Shih-Chieh Huang  •  School of Natural Sciences, University of California, Merced, CA, USA Yong-Min Huh  •  Department of Radiology, Yonsei University, Seoul, Korea Kazuhiko Ishihara  •  Department of Materials Engineering, Center for NanoBio Integration, The University of Tokyo, Tokyo, Japan Gopal Iyer  •  Department of Chemistry and Biochemistry, University of California at Los Angeles, Los Angeles, CA, USA Xiqun Jiang  •  Laboratory of Mesoscopic Chemistry, Department of Polymer Science and Engineering, College of Chemistry and Chemical Engineering, Nanjing University, Nanjing, People’s Republic of China Noriho Kamiya  •  Department of Applied Chemistry, Graduate School of Engineering, Kyushu University, Fukuoka, Japan Stefan Kaufmann  •  Laboratory for Surface Science and Technology, Swiss Federal Institute of Technology Zurich (ETH Zurich), Zurich, Switzerland Jörg Kressler  •  Institute of Chemistry, Martin Luther University, Halle Wittenberg, Germany Joanna Kubler-Kielb  •  Program on Developmental and Molecular Immunity, Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, MD, USA Karthik Kumar  •  Laboratory for Surface Science and Technology, Swiss Federal Institute of Technology Zurich (ETH Zurich), Zurich, Switzerland

Contributors

Thomas Kurpiers  •  Ascendis Pharma GmbH, Heidelberg, Germany Minseok Kwak  •  Department of Polymer Chemistry, The Zernike Institute for Advanced Materials, University of Groningen, Groningen, The Netherlands Kwangyeol Lee  •  Department of Chemistry, Korea University, Seoul, Korea Edward A. Lemke  •  Structural and Computational Biology Unit, European Molecular Biology Laboratory, Heidelberg, Germany Edith H.M. Lempens  •  Laboratory of Chemical Biology, Department of Biomedical Engineering, Eindhoven University of Technology, Eindhoven, The Netherlands Eun-Kyung Lim  •  Department of Chemical and Biomolecular Engineering, Yonsei University, Seoul, Korea Shyh-Chyang Luo  •  Yu Initiative Research Unit, RIKEN Advanced Science Institute, Saitama, Japan John H.T. Luong  •  Biotechnology Research Institute, National Research Council Canada, Montreal, QC, Canada Yong Ma  •  Department of Chemistry, Cleveland State University, Cleveland, OH, USA Karsten Mäder  •  Institute of Pharmacy, Martin Luther University, Halle Wittenberg, Germany Khaled A. Mahmoud  •  Biotechnology Research Institute, National Research Council Canada, Montreal, QC, Canada Keith B. Male  •  Biotechnology Research Institute, National Research Council Canada, Montreal, QC, Canada Julio A. Martinez  •  School of Natural Sciences, University of California, Merced, CA, USA Andreas Marx  •  Department of Chemistry, Konstanz Research School Chemical Biology, University of Konstanz, Konstanz, Germany Andrea Masotti  •  Gene Expression – Microarrays Laboratory, IRCCS - Bambino Gesù Children’s Hospital, Rome, Italy Maarten Merkx  •  Laboratory of Chemical Biology, Department of Biomedical Engineering, Eindhoven University of Technology, Eindhoven, The Netherlands Anna Mero  •  Department of Pharmaceutical Sciences, University of Padova, Padova, Italy Albert Meyer  •  Institut des Biomolécules Max Mousseron, Université Montpellier, Montpellier, France Nipun Misra  •  School of Natural Sciences, University of California, Merced, CA, USA Bikash Mondal  •  Department of Chemistry, The City University of New York at the College of Staten Island, Staten Island, NY, USA Henning D. Mootz  •  Institut für Biochemic, University of Muenster, Münster, Germany François Morvan  •  Institut des Biomolécules Max Mousseron, Université Montpellier, Montpellier, France Andrew J. Musser  •  Department of Polymer Chemistry, The Zernike Institute for Advanced Materials, University of Groningen, Groningen, The Netherlands

xiii

xiv

Contributors

Peter E. Nielsen  •  Department of Cellular and Molecular Medicine, The Panum Institute, Faculty of Health Sciences, University of Copenhagen, Copenhagen, Denmark Kazuki Nishizawa  •  Department of Materials Engineering, Center for NanoBio Integration, The University of Tokyo, Tokyo, Japan Aleksandr Noy  •  School of Natural Sciences, University of California, Merced, CA, USA Francesco Pampaloni  •  Physical Biology Group Institute for Cell Biology and Neurosciences Frankfurt Institute for Molecular Life Sciences (FMLS) , Goethe University Frankfurt, Frankfurt am Main, Germany Marta Pasek  •  Structural Glycobiology Section, Center for Cancer Research Nanobiology Program, National Cancer Institute, Frederick, MD, USA Gianfranco Pasut  •  Department of Pharmaceutical Sciences, University of Padova, Padova, Italy Hui Peng  •  Key Laboratory of Polar Materials and Devices, College of Information Science and Technology, East China Normal University, Shanghai, China Sónia Pérez-Rentero  •  Institute for Research in Biomedicine, IQAC-CSIC, CIBER-BBN Networking Centre on Bioengineering, Biomaterials and Nanomedicine, Barcelona, Spain Markus Pietzsch  •  Institute of Pharmacy, Martin Luther University, Halle Wittenberg, Germany Tilo Pompe  •  Leibniz Institute of Polymer Research Dresden, Max Bergmann Center of Biomaterials Dresden, Dresden, Germany Gwladys Pourceau  •  Institut des Biomolécules Max Mousseron, Université Montpellier, Montpellier, France Pradman K. Qasba  •  Structural Glycobiology Section, Center for Cancer Research Nanobiology Program, National Cancer Institute, Frederick, MD, USA Krishnaswami Raja  •  Department of Chemistry, The City University of New York at the College of Staten Island, Staten Island, NY, USA Boopathy Ramakrishnan  •  Structural Glycobiology Section, Center for Cancer Research Nanobiology Program, National Cancer Institute, Frederick, MD, USA; Basic Science Program, SAIC-Frederick, Inc., Frederick, MD, USA Erik Reimhult  •  Laboratory for Surface Science and Technology, Swiss Federal Institute of Technology Zurich (ETH Zurich), Zurich, Switzerland Olivier Renaudet  •  Département de Chimie Moléculaire, Université de Grenoble, Grenoble, France Katrin Salchert  •  Lausitz University of Applied Sciences, Senftenberg, Germany Rupa R. Sawant  •  Department of Pharmaceutical Sciences, Center for Pharmaceutical Biotechnology and Nanomedicine, Northeastern University, Boston, MA, USA Fwu-Shan Sheu  •  NUSNNI-Nanocore Institute, National University of Singapore, Singapore, Singapore; Department of Electrical and Computer Engineering, National University of Singapore, Singapore, Singapore Wei Shi  •  Department of Chemistry, The City University of New York at the College of Staten Island, Staten Island, NY, USA

Contributors

xv

Takehiko Shiraishi  •  Department of Cellular and Molecular Medicine, The Panum Institute, Faculty of Health Sciences, University of Copenhagen, Copenhagen, Denmark Mariusz Skwarczynski  •  School of Chemistry and Molecular Biosciences, The University of Queensland, St. Lucia, QLD, Australia Matthew D. Smith  •  Department of Chemistry, The University of Tennessee, Knoxville, TN, USA Xue-Long Sun  •  Department of Chemistry, Cleveland State University, Cleveland, OH, USA Madoka Takai  •  Department of Materials Engineering, Center for NanoBio Integration, The University of Tokyo, Tokyo, Japan Montserrat Terrazas  •  Institute for Research in Biomedicine, IQAC-CSIC, CIBER-BBN Networking Centre on Bioengineering, Biomaterials and Nanomedicine, Barcelona, Spain Thomas J. Tolbert  •  Department of Chemistry, Indiana University, Bloomington, IN, USA Vladimir P. Torchilin  •  Department of Pharmaceutical Sciences, Center for Pharmaceutical Biotechnology and Nanomedicine, Northeastern University, Boston, MA, USA Istvan Toth  •  School of Chemistry and Molecular Biosciences, The University of Queensland, St. Lucia, QLD, Australia Jadranka Travas-Sejdic  •  Polymer Electronics Research Centre, Department of Chemistry, The University of Auckland, Auckland, New Zealand Jean-Jacques Vasseur  •  Institut des Biomolécules Max Mousseron, Université Montpellier, Montpellier, France Francesco M. Veronese  •  Department of Pharmaceutical Sciences, University of Padova, Padova, Italy Qianli Wang  •  Department of Chemistry, Wayne State University, Detroit, MI, USA Samuel H. Weisbrod  •  Department of Chemistry, Konstanz Research School Chemical Biology, University of Konstanz, Konstanz, Germany Shimon Weiss  •  Departments of Chemistry & Biochemistry and Physiology, University of California at Los Angeles, Los Angeles, CA, USA; California NanoSystems Institute, Los Angeles, CA, USA Carsten Werner  •  Leibniz Institute of Polymer Research Dresden, Max Bergmann Center of Biomaterials Dresden, Dresden, Germany; Center of Regenerative Therapies Dresden, Dresden, Germany Nathan P. Westcott  •  Department of Chemistry, The Carolina Center for Genome Sciences, The University of North Carolina, Chapel Hill, NC, USA Wei Wu  •  Laboratory of Mesoscopic Chemistry, Department of Polymer Science and Engineering, College of Chemistry and Chemical Engineering, Nanjing University, Nanjing, People’s Republic of China Junpeng Xiao  •  Department of Chemistry, Indiana University, Bloomington, IN, USA Jianmin Xu  •  Department of Chemistry and Biochemistry, University of California at Los Angeles, Los Angeles, CA, USA

xvi

Contributors

Muhammad N. Yousaf  •  Department of Chemistry, The Carolina Center for Genome Sciences, The University of North Carolina, Chapel Hill, NC, USA Hsiao-hua Yu  •  Yu Initiative Research Unit, RIKEN Advanced Science Institute, Saitama, Japan Qin Zhang  •  Department of Occupational Health, West China School of Public Health, Sichuan University, Chengdu, People’s Republic of China Hailong Zhang  •  Department of Chemistry, Cleveland State University, Cleveland, OH, USA Shengchao Zhu  •  Research Center for Eco-environmental Sciences, Chinese Academy of Sciences, Beijing, People’s Republic of China Jan Zimmermann  •  Department of Polymer Chemistry, The Zernike Institute for Advanced Materials, University of Groningen, Groningen, The Netherlands

Part I Protein Conjugates

Chapter 1 Site-Specific Labeling of Proteins for Single-Molecule FRET Measurements Using Genetically Encoded Ketone Functionalities Edward A. Lemke Abstract Studies of protein structure and function using single-molecule fluorescence resonance energy transfer (smFRET) benefit dramatically from the ability to site-specifically label proteins with small fluorescent dyes. Genetically encoding the unnatural amino acid (UAA) p-acetylphenylalanine is an efficient way to introduce commercially available fluorescent tags with high yield and specificity. This protocol describes the expression in Escherichia coli of proteins containing this UAA in response to the amber stop codon TAG. Proteins were purified with high yield and subsequently labeled with the hydroxylamine derivative of Alexa Fluor® 488 functioning as a fluorescent donor dye. The proteins were then labeled via maleimide coupling chemistry at a unique cysteine with the acceptor dye Alexa Fluor® 594 to yield a dual-labeled protein ready for subsequent smFRET observation. Key words: Amber stop codon, Genetically encoded, Synthetase, Unnatural amino acid, Oxime ligation, Maleimide coupling, Single-molecule, FRET, Protein dynamics, Site-specific labeling

1. Introduction Labeling proteins with small fluorescent dyes has long been an established tool to make molecular structure and function visible. Among the plethora of different fluorescence techniques, the observation of fluorescence resonance energy transfer (FRET) between a fluorescent donor dye and a fluorescent acceptor dye has emerged as a widely applied tool in biology. FRET occurs if the emission spectrum of a donor dye overlaps with the absorption spectrum of a proximal acceptor dye. Since the efficiency of energy transfer is distance-dependent, an exact measurement of the FRET efficiency can be used to study distances between

Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_1, © Springer Science+Business Media, LLC 2011

3

4

Lemke

20 and 100 Å. Due to the pioneering work of Ha, Deniz and coworkers, it has also become possible to observe fluorescence energy transfer on the single-molecule level using highly sensitive detection methods (1, 2). In order to turn such a single-molecule FRET (smFRET) measurement into a structural biology tool to study molecular function, the labeling sites must be precisely known. Recombinantly expressed proteins are typically composed of 20 naturally occurring amino acids (AA). Of those AA, lysine and cysteine have chemical handles (the side-chain amine group of lysine and the thiol group of cysteine) for which common labeling strategies exist to react them efficiently with commercially available dye derivatives that are suitable for smFRET studies. Due to the rare occurrence of cysteine in most proteins, many smFRET studies on proteins entail encoding two surface accessible cysteines into the protein sequence (at labeling sites X and Y). The cysteines are then reacted with maleimide-reactive dyes, typically leading to randomly labeled species, as reaction sites X and Y are essentially chemically indistinguishable. Such a random labeling strategy creates a heterogeneous sample containing a mixture of donor (D) and acceptor (A) dye species attached at positions X and Y (such as XD/YA, XA/YD). If, for example, XA and YA have different photophysical properties, a FRET measurement could yield distinct signals for each species. This can be detrimental for data interpretation and can compromise the unique capability of single-molecule studies to look beyond the average and detect subpopulations, which has greatly contributed to the technique’s success in yielding an unbiased view of even complex biological processes. As a consequence, a randomly labeled, intrinsically heterogeneous sample can cause severe downstream problems in data analysis and interpretation of biologically relevant subpopulations. Genetically encoding the unnatural amino acid (UAA) p-acetylphenylalanine (pAcPhe) (Fig.  1a) using amber suppression technology has created an orthogonal route to achieve sitespecific dual-labeling of proteins with dyes suitable for single-molecule observation (3, 4). Here, the expression host, E. coli, was engineered with an additional aminoacyl-tRNA synthetase/suppressor tRNA pair (aaRS/tRNACUA). This pair was derived from Methanocaldococcus jannaschii and modified in such a way that the synthetase specifically recognizes pAcPhe and aminoacetylates only its cognate tRNACUA (5). This leads to incorporation of this specific UAA during the translation process whenever the TAG codon occurs. E. coli (and many other organisms) tolerates this reprogramming of one of its rare codons. Transforming a plasmid encoding for the pAcPheaaRS/tRNACUA pair into E. coli thus extends the capability of the host to genetically encode the 21st amino acid pAcPhe. Furthermore, the system is

Site-Specific Labeling of Proteins for Single-Molecule FRET Measurements

a

c

b

O

ara

ClAmp

COOH

d 34 26 17 10

pAcPhe

crude

M

+

POI

tRNACUA

H2N

5

pure

-

+

aaRS

pAcPhe

ara aaRS

Amp

Fig. 1. (a) Structure of p-acetylphenylalanine (pAcPhe). (b) Plasmid pEVOLpAcPhe encoding ClAmp resistance, tRNACUA, and two copies of pAcPheaaRS (one under control of an l-arabinose-inducible promoter). (c) pBAD expression vector encoding Amp resistance and the protein of interest (POI) (T4 Lysozyme in this protocol) under control of an l-arabinose-inducible promoter. (d) Coomassie-stained SDS-PAGE gel showing expression of T4 Lysozyme 38TAG 157C in the presence (+) and absence (−) of pAcPhe after the first (crude) purification step, as well as the final purified 38pAcPhe 157C T4 Lysozyme protein product.

orthogonal to the existing translational machinery in E. coli, i.e., it does not cross-react or interfere with any of the natural aaRS/tRNA pairs. Expression of a protein of interest (POI) using this 21st UAA only requires a few additional steps compared to standard expression procedures. First, a TAG mutation is introduced by standard site-directed mutagenesis at the desired labeling site and cloned into a suitable expression vector. For transformation, this vector has to carry a resistance marker that is compatible with the expression host as well as with the plasmid harboring the pAcPheaaRS/ tRNACUA pair (termed pEVOLpAcPhe; see Fig. 1b) (5). The vector pEVOLpAcPhe carries a gene encoding for chloramphenicol (ClAmp) resistance and was optimized for high yield expression in E. coli. This plasmid encodes two genes for pAcPheaaRS; because one of the synthetase genes is under direct control of an arabinose-inducible promoter, high yield expression of a protein containing pAcPhe requires induction with l-arabinose. The UAA pAcPhe is membrane permeable and thus can be made available for the translation process by simply adding it to the growth medium prior to induction. As the actual incorporation of UAAs into proteins is carried out by the host translation machinery in vivo, the supplied UAA does not need to be stereochemically pure, as pAcPheaaRS recognizes only l-amino acids. And finally, as amber codon suppression is not necessarily 100% efficient, the POI should contain a C-terminal purification handle (such as a histidine tag), so that it can be easily discriminated during the purification process from truncated mutants. The following protocol describes the efficient incorporation of pAcPhe into a single-cysteine mutant of T4 Lysozyme. While protein expression differs from protein to protein, most steps specific

6

Lemke O

SH SO3

SO3

+

O

H2N

NH2

oxime ligation

C Alexa488

H2N

=

_

O

O HN

O

O

HN

H2N

O

O

Alexa488

O N

SH

CH3 H3C

N

CH3

_

=

COH

CH2SO3

Alexa594

_

CH2SO3

O O

HN O

O

N

O

O N

CH3

N

H3C

O

maleimide coupling

Alexa488

CH3 O

N

Alexa594

O N O S

Fig. 2. T4 Lysozyme 38pAcPhe 157C is first labeled with Alexa Fluor® 488 hydroxylamine using oxime ligation at pH 4, and then Alexa Fluor® 594 C5 maleimide is coupled to the cysteine residue at pH 7.4.

to incorporation of pAcPhe will be common to all proteins, as pointed out in the text. Subsequently, the cysteine and the pAcPhe are site-specifically labeled with a commercially available smFRET dye pair (Fig. 2).

Site-Specific Labeling of Proteins for Single-Molecule FRET Measurements

7

2. Materials 1. E. coli TOP10 electrocompetent cells (Invitrogen Corp., Carlsbad, CA). 2. Glycerol. 3. Luria–Bertani (LB) growth medium. 4. LB agar plates with 50 mg/L ampicillin and 33 mg/L ClAmp. 5. 20% (w/v) l-arabinose, filter-sterilized through a 0.2-mm syringe filter. 6. pAcPhepEVOL (available from Prof. Peter G. Schultz, The Scripps Research Institute, San Diego, CA). 7. 157Cys T4 Lysozyme-6His gene cloned into pBAD-MCS (Invitrogen) using NcoI and KpnI restriction sites. 8. QuikChange® site-directed mutagenesis kit (Agilent Technologies, Santa Clara, CA) or other site-directed mutagenesis kit. 9. Lysis buffer: 20  mM sodium phosphate–HCl, pH 7.6, 500 mM NaCl, 1 mM imidazole, 1 mM PMSF, complete protease inhibitor cocktail (e.g., Roche Diagnostics Corporation, Indianapolis, IN). 10. Mono S ion exchange chromatography column (GE Healthcare, Piscataway, NJ). 11. Fast protein liquid chromatographic (FPLC) equipment. 12. Wash buffer: 20  mM sodium phosphate–HCl, pH 7.6, 500 mM NaCl, 10 mM imidazole, and 1 mM PMSF. 13. Elution buffer: 20  mM sodium phosphate–HCl, pH 7.6, 500 mM NaCl, 500 mM imidazole, and 1 mM PMSF. 14. Mono S buffer A: 50 mM MES–HCl, pH 6.3, 50 mM NaCl, and 5 mM b-mercaptoethanol. 15. Mono S buffer B: 50 mM MES–HCl, pH 6.3, 500 mM NaCl, and 5 mM b-mercaptoethanol. 16. Oxime labeling buffer: 50 mM sodium acetate–HCl, pH 4.0, 150 mM NaCl, and 4 M guanidine HCl. 17. Maleimide labeling buffer: 20 mM sodium phosphate–HCl, pH 7.4, 150 mM NaCl, and 4 M guanidine HCl. 18. Ni-NTA Superflow resin (Qiagen, Valencia, CA). 19. Alexa Fluor® 488 C5-aminooxyacetamide, bis(triethy-lammonium) salt (Alexa Fluor® 488 hydroxylamine) (Invitrogen) (see Note 1). 20. Alexa Fluor® 594 C5 maleimide (Invitrogen). 21. 1  M pAcPhe HCL stock solution. Filter-sterilize through a 0.2-mm syringe filter (e.g., available from Synchem, IL, USA).

8

Lemke

22. Dimethylsulfoxide (anhydrous). 23. Acetonitrile (anhydrous). 24. Dithiothreitol (DTT). 25. Polypropylene chromatography column with filter (Qiagen). 26. NuPage denaturing SDS-PAGE gel (4–12% Bis–Tris Gel) (Invitrogen). 27. SDS-PAGE MOPS running buffer. 28. SDS-PAGE loading dye. 29. SDS-PAGE molecular weight markers. 30. Centricon® centrifugal filter units (10  kDa MWCO) (Millipore, Billerica, MA) or 10 kDa MWCO dialysis tubes. 31. UV–visible spectrophotometer. 32. Mass spectrometer.

3. Methods High-yielding protein expression greatly simplifies subsequent protein purification. For most smFRET experiments, the final purity of the sample is pivotal, because otherwise impurities could get labeled as well and, therefore, contribute to fluorescent background. The pAcPheaaRS/tRNACUA pair is highly efficient in selective incorporation of pAcPhe in E. coli and several proteins have already been expressed in milligram quantities from a 1-L culture (5). Misincorporation of a naturally occurring amino acid instead of pAcPhe would yield protein that does not contain a ketone handle, and that consequently is missing a labeling site. Such impurities are highly undesirable, and the described strategy for UAA incorporation was developed to allow for the preparation of pure modified protein. The unique side-chain ketone functionality of pAcPhe is inert and nontoxic to E. coli, but reacts efficiently with hydroxylamine dye derivatives by forming a stable oxime bond. 3.1. Transformation of E. coli and Starting Culture

1. Perform site-directed mutagenesis (e.g., a QuikChange® ­reaction) to mutate a codon in your protein sequence to TAG. In this example, the codon for amino acid 38 in the 157C T4 Lysozyme was mutated to TAG and confirmed by DNA sequencing. This generates the pBAD plasmid encoding 38TAG 157C T4 Lysozyme (see Note 2). 2. Transform E. coli TOP10 cells with the pAcPhepEVOL (ClAmp resistance, inducible with l-arabinose) and pBAD-38TAG 157C T4 Lysozyme (ampicillin resistant, inducible with l-arabinose) plasmids (see Note 3).

Site-Specific Labeling of Proteins for Single-Molecule FRET Measurements

9

3. Spread transformed colonies on LB agar plates with 50 mg/L ampicillin and 33 mg/L ClAmp and grow overnight at 37°C (from here on, this antibiotic mixture is referred to as Amp/ ClAmp). Pick an individual colony and grow overnight in LB medium containing Amp/ClAmp at 37°C. Make a 30% glycerol stock from this overnight culture and store at −80°C. Start all further experiments from this glycerol stock by scratching off a small amount and growing an overnight culture in double-selective medium (Amp/ClAmp) (see Note 4). 3.2. Protein Expression

1. Inoculate two 2-L shake flasks (each containing 500 mL liquid LB medium with Amp/ClAmp) with 5 mL from an overnight seed culture and allow cells to grow at 37°C with constant shaking. Add pAcPhe to a final concentration of 1  mM to one of the flasks (+) at an optical density of OD600 = 0.2–0.3. The other flask serves as a control (−). 2. Let the cells continue to grow until mid-log phase (OD600 = 0.4–0.6), and then induce recombinant protein expression by adding l-arabinose at a final concentration of 0.02% (w/v). 3. Transfer cells to a 30°C shaker and shake overnight for 16 h (see Note 5). Pellet cells by centrifugation at 4,000 × g for 20 min. Discard the supernatant and freeze the cell pellet at −80°C.

3.3. Protein Purification (Part I)

1. Resuspend the cell pellet in 3  volumes of lysis buffer, and then lyse the cells using sonication. 2. Centrifuge the lysate at 16,000 × g for 60  min. Collect the supernatant and incubate with 2 mL of Ni-NTA beads (preequilibrated in lysis buffer) for 2  h at 4°C (under constant mixing using a nutator agitation device). 3. Collect the Ni-NTA resin by passing the lysate through a polypropylene column so that the Ni-NTA beads form a bed above the filter at the bottom of the column. Wash the column with 20 bed volumes of wash buffer. 4. Elute the target protein from the column using three bed volumes of elution buffer.

3.4. Expression Test Using SDS-PAGE

1. Take a sample from the (+) and the (−) eluates and run on an SDS-PAGE gel. An exemplar gel is shown in Fig. 1. In the (+) pAcPhe sample lane, an enriched product at the molecular weight of T4 Lysozyme is evident. For comparison, the (−) pAcPhe sample lane contains the negative control where no pAcPhe was added. Besides minor impurities in both samples, substantial expression of T4 Lysozyme is only visible in the (+) UAA sample. This is a strong indication that the incorporation of pAcPhe was successful (see Notes 6 and 7).

10

Lemke

3.5. Protein Purification (Part II)

Only the (+) UAA expression sample crudely purified in Subheading 3.3 will contain pAcPhe. Therefore, only the pAcPhecontaining sample (+) is used for further purification. 1. Transfer the sample to a Centricon® concentration device with a molecular weight cutoff of 10 kDa, concentrate and wash three times with Mono S buffer A (alternatively, use dialysis for buffer exchange). At this point, a white precipitate forms (see Note 8). Discard the white precipitate material since it originates from protein that is not soluble at pH 6.3 and does not contain the target protein T4 Lysozyme. 2. Concentrate the protein sample to 1  mL and load onto a Mono S column at a flow rate of 0.5 mL/min using an FPLC purification system. After a stable baseline is reached, apply a salt gradient at a flow rate of 1 mL/min (from 100% Mono S buffer A to 100% Mono S buffer B over 30 min). T4 Lysozyme typically elutes around a salt concentration of 60% Mono S buffer B. Analyze the collected FPLC fractions on SDS-PAGE gels (see Note 9). 3. Pool the fractions containing purified T4 Lysozyme together and concentrate in a Centricon® filter device. At this point, exchange the buffer using three repeated washes with oxime labeling buffer. Measure the concentration of protein on a UV–visible spectrophotometer (see Note 10). The yield of 38pAcPhe 157C T4 Lysozyme is typically >15  mg from a 0.5-L expression culture. Confirm successful purification by mass spectrometry (see Note 7).

3.6. Oxime Ligation of T4 Lysozyme Containing pAcPhe with Axela 488 Hydroxylamine Dye

1. Mix 50 mL of 200 mM 38pAcPhe 157C T4 Lysozyme protein with 50  mL of 1  mM Alexa Fluor® 488 hydroxylamine dye (i.e., fivefold molar excess of dye) in pH 4 oxime labeling buffer (see Note 11). Allow the reaction to proceed for 36 h at 37°C in the dark. 2. Wash the labeled protein three times with maleimide labeling buffer. 3. Efficient labeling is verified by SDS-PAGE (see Note 9), UV–visible spectrophotometry (see Note 10), and mass spectrometry (see Note 7).

3.7. Aliquoting Alexa Fluor ® 594 C5 Maleimide for Multiple Reactions

Maleimide-reactive dyes are typically supplied commercially by companies in 1-mg quantities. However, for most single-molecule FRET studies a couple of nanomoles of dual-labeled protein is sufficient, and thus it is desirable to aliquot the maleimide dye for separate reactions. In contrast to most hydroxylamine dye derivates, the maleimide functional group is not stable in water. To allow for multiple reactions out of a single 1-mg sample of Alexa Fluor® 594 C5 maleimide dye, carefully aliquot the dye

Site-Specific Labeling of Proteins for Single-Molecule FRET Measurements

11

under inert gas (either provided by constant nitrogen or argon flow, or by using a glove box). 1. Solubilize 1 mg of Alexa Fluor® 594 C5 maleimide in anhydrous acetonitrile (see Note 12) 2. Aliquot the Axela Fluor® 594 C5 maleimide solution, e.g., in 25-nmol fractions into dry microcentrifuge tubes 3. Lyophilize the aliquots and store at −80°C. 3.8. Coupling of Alexa Fluor ® 594 C5 Maleimide Dye to the Cysteine Residue of T4 Lysozyme

1. Wash the Alexa Fluor® 488-labeled protein three times with maleimide buffer + 10 mM DTT, and then subsequently wash three more times with maleimide labeling buffer (see Note 13). 2. Set up the following labeling reaction using a protein:Alexa Fluor® 594 dye labeling ratio of 1:1.5. Prepare a solution containing 10  nmol of protein in 80  mL of maleimide labeling buffer. Next, solubilize 15  nmol of Alexa Fluor® 594 C5 maleimide in 20  mL of DMSO and pipette the dye solution into the protein solution while gently vortexing the protein in a microcentrifuge tube. This instantaneous mixing procedure prevents any formation of precipitate. Allow the reaction to proceed for 1 h at room temperature (or overnight at 4°C) and then quench the reaction by adding DTT to a final concentration of 50 mM or proceed immediately to the next step. 3. Exchange the protein into water either by dialysis or by repeated washes in a Centricon® device and subsequently lyophilize for long-term storage (in the dark) at −80°C. 4. Analyze the final sample by using SDS-PAGE (see Note 9), UV–visible spectrophotometry (see Note 10), and mass spectrometry (see Note 7). Typically, site-specific dual-labeling efficiencies are larger than 90% for each labeling site. The lyophilized T4 Lysozyme will be resolubilized best in a denaturing buffer, at which point it is now ready for subsequent single-molecule FRET studies, such as those described in Lemke et al. (3) or Brustad et al. (4).

4. Notes 1. To prepare 1 mM Alexa Fluor® 488 hydroxylamine solution, dissolve 1 mg of dye in pH 4 oxime labeling buffer. This solution can be stored frozen in the dark at −20°C and thawed prior to use. 2. Ideally, only surface-accessible mutants that are unlikely to interfere with protein folding, structure, and function should be chosen as amber suppression sites. Furthermore, it is

12

Lemke

imperative that the stop codon used for termination of the protein sequence is not TAG, but either TAA or TGA. 3. If expression problems occur, make a new glycerol stock of transformed E. coli TOP10 cells (Subheading  3.1, step 2). Further, the pEVOL plasmid should be analyzed for unwanted mutations specifically in the tRNA and aaRS coding regions. 4. Protein expression conditions typically need to be adapted for every given POI. The pAcPhepEVOL system is compatible with most commercially available E. coli hosts. Only a few things need to be considered: (a) the antibiotic selection markers of the plasmids used must be unique and compatible with each other and with the host; (b) the host must support expression with l-arabinose; and (c) it is advisable that the UAA pAcPhe is added 30 min prior to induction. This gives sufficient time for the tRNACUA to become charged with pAcPhe. pAcPhe is a very stable UAA and can in principle also be added to the medium right from the beginning. 5. Expression at 30°C was chosen for this particular protein. Other growth conditions and temperatures can be used for other proteins. Different expression conditions can also affect amber suppression. However, stable, high-yielding amber suppression using pAcPheaaRS/tRNACUA has been observed across a wide spectrum of common expression conditions. 6. The (+)/(−) pattern is a strong indication for successful incorporation of the UAA pAcPhe. However, the following points may be considered if problems occur. The presence of a band in the (−) pAcPhe lane does not necessarily report a failure of the experiment. In the presence of pAcPhe, the binding pocket is occupied by pAcPhe due to the high affinity of the synthetase (pAcPheaaRS) for this specific UAA. In the absence of UAA, it is possible for the binding pocket of the synthetase to be filled by an unwanted substrate that is then charged to the tRNACUA. In such a case, both the (+) and the (−) lanes will show expression of the target protein, but pAcPhe is only incorporated in the (+) UAA case. The (+)/(−) band ratio is also dependent on various factors, including type of growth medium, growth conditions, and concentration of pAcPhe in the medium. Increasing the pAcPhe concentration to 5 mM can increase the yield of the target protein incorporating pAcPhe. Independent of the appearance of the (+)/(−) ratio, high-resolution mass spectrometry must be used to verify sufficient (>90%) incorporation of pAcPhe in the (+) UAA experiment (see Note 10). If the (+)/(−) ratio is not satisfactory, the following steps can be taken: perform protein expression in different growth media (such as TYT, TB) at different temperatures and vary the expression time. As the pAcPheaaRS/ tRNACUA pair is evolved from a wild-type tyrosine synthetase,

Site-Specific Labeling of Proteins for Single-Molecule FRET Measurements

13

phenylalanine (F) and tyrosine (Y) incorporation are the most likely types of misincorporations. Often, expression in minimal media (lacking Y/F) can dramatically increase specificity. 7. Most essential is the confirmation of successful incorporation of pAcPhe at the specific protein site using high-resolution mass spectrometry, such as electrospray ionization or matrixassisted laser desorption/ionization. The expected mass from the protein can be calculated using standard protein mass calculators (such as the peptide mass calculator available via the http:// www.expasy.ch webpage). Because most calculators do not easily allow for modified unnatural residues, the most straightforward way to do this is to substitute a tyrosine (Y) at the position of the expected pAcPhe incorporation site and add 26 Da, which reflects the mass increase when substituting the tyrosine hydroxyl with a ketone functionality. To remove further uncertainties in the mass calculation due to, e.g., protein modifications (such as acetylation), one should also compare the molecular weight to a sample encoding tyrosine at the mutation site. Additionally, the protein sample can be digested with trypsin and peptide masses be analyzed. In our hands, the exact mass of most of the unmodified peptides was confirmed, but in our specific case the peptide containing pAcPhe was not detected (SPXLNAAK with X = pAcPhe; expected mass (M + H) = 889.46 Da). However, after the first labeling reaction, the same mass spectrometric analyses were repeated and a mass increase of the whole protein of 673 Da was verified, confirming labeling with Alexa Fluor® 488 hydroxylamine. The labeled peptide SPXLNAAK could now also be identified (with X = pAcPhe + Alexa Fluor® 488 hydroxylamine; (M + H) = 1,562 Da). After the second labeling reaction with Alexa Fluor® 594 C5 maleimide, a further mass increase of the whole protein of 886  Da was also confirmed. After tryptic digestion, this labeled peptide was also identified (TGCWDAYK + Alexa Fluor® 594; (M + H) = 1,830 Da). 8. The Mono S purification step is specific for T4 Lysozyme. Other proteins may require a different purification strategy to yield pure protein. Make sure that the molecular weight cutoff of the centrifugal filter or dialysis devices is compatible with the molecular weight of the target protein. 9. SDS-PAGE analysis is used to confirm the expression of the POI. When using a fluorescence scanner, this can also be used to rapidly confirm fluorescence labeling. For detecting a fluorescently-labeled band, the gel should be run in the dark and analyzed on a fluorescence scanner prior to Coomassie staining (otherwise Coomassie can quench the fluorescence signal). After the fluorescence scan, the gel can be stained with Coomassie. The fluorescence image should show a band

14

Lemke

clearly overlaying the Coomassie-stained band. Additional fluorescence bands can pinpoint labeled impurities in the gel (fluorescence detection is very sensitive to even minor impurities) or free dye still present in the sample. 10. After the purification of the protein, the concentration of the protein should be measured using a standard UV–visible spectrophotometer. If the signal-to-noise of the UV–visible absorbance spectrum is not satisfactory, increase the concentration of the protein. Furthermore, exchanging the protein into a low-salt buffer (such as water) can yield a better signal, as salts often contribute to a high level of background absorption in the UV range. The total concentration of pure protein can be determined by measuring the absorbance A280 at the wavelength of 280 nm on a UV–visible spectrophotometer. Using the extinction coefficient of e280, T4L ~ 26,000 cm−1 M−1, the concentration of pure protein can be calculated according to c protein = A280 / e 280,protein . After each labeling reaction, the labeling efficiency cdye/cprotein is determined by calculating the concentration of fluorescent dye and the concentration of protein in the sample as given by c dye = Adye / e dye and c protein = (A280 - corr280 ·Adye ) / e 280,protein , respectively, where edye is the extinction coefficient for the dye at its maximal absorption wavelength, lmax. Most fluorescent dyes also absorb light at 280 nm, which has to be corrected for when calculating the protein concentration using the corr280 parameter. Typically, the extinction coefficient for a given dye, as well as the correction factor (corr280), is available from the manufacturer’s manual. For the specific dyes used in this example protocol, the values are as follows: (1) Alexa Fluor® 488: e495 ~ 71,000 cm−1 M−1, lmax = 495 nm, and corr280 = 0.11; and (2) Alexa Fluor® 594: e594 ~ 90,000 cm−1 M−1, lmax = 590 nm, and corr280 = 0.56. Typical labeling efficiencies obtained for this example protocol are >90%. As the two labeling sites are orthogonal to each other, the order of labeling can in principle be reversed, i.e., one may first perform the maleimide reaction and then perform the oxime ligation. 11. The oxime ligation is mainly dependent on pH, concentration of educts, time, and temperature. Labeling efficiencies can be increased by using higher concentration of educts (i.e., protein and/or dye), incubation at 37°C for a longer time period, or raising the temperature to, e.g., 65°C. The optimum of the reaction is around pH 4. Thus, if a higher pH is needed, then the temperature, concentration, and incubation time should be optimized accordingly. Furthermore, the additional catalysts can also be used (6). 12. Many maleimide-reactive dyes are not readily soluble in low amounts of acetonitrile. Increasing the amount of acetonitrile

Site-Specific Labeling of Proteins for Single-Molecule FRET Measurements

15

as well as extensive vortexing and sonication on ice can assist solubility. The dye must not be exposed to ambient air or water. 13. The presence of residual thiol-containing contaminants is the most typical cause for low labeling efficiencies. Furthermore, b-mercaptoethanol used in previous purification steps can block the thiol group of the free cysteine. To avoid this, extensively wash or dialyze the protein prior to labeling with buffer containing 10 mM DTT, followed by extensive washes in fresh, deoxygenated thiol-free buffer to remove even trace amounts of thiol-containing compounds.

Acknowledgments I thank all members in the Deniz and Schultz laboratories at The Scripps Research Institute, especially Dr. Brustad for the good collaborations. The critical proofreading of this protocol by Dr. VanDelinder is also very much appreciated. Finally, I want to thank my laboratory members for their productive discussions, and the EMBL and the Emmy Noether Program of the DFG for funding my laboratory. References 1. Ha, T., Enderle, T., Ogletree, D. F., Chemla, D. S., Selvin, P. R., and Weiss, S. (1996) Probing the interaction between two single molecules: fluorescence resonance energy transfer between a single donor and a single acceptor. Proc. Natl. Acad. Sci. USA 93, 6264–6268. 2. Deniz, A. A., Dahan, M., Grunwell, J. R., Ha, T., Faulhaber, A. E., Chemla, D. S., Weiss, S., and Schultz, P. G. (1999) Single-pair fluorescence resonance energy transfer on freely diffusing molecules: observation of Forster distance dependence and subpopulations. Proc. Natl. Acad. Sci. USA 96, 3670–3675. 3. Brustad, E. M., Lemke, E. A., Schultz, P. G., and Deniz, A. A. (2008) A general and efficient method for the site-specific dual-labeling of

proteins for single molecule fluorescence resonance energy transfer. J. Am. Chem. Soc. 130, 17664–17665. 4. Lemke, E. A., Gambin, Y., Vandelinder, V., Brustad, E. M., Liu, H.-W., Schultz, P. G., Groisman, A., and Deniz, A. A. (2009) Microfluidic Device for Single-Molecule Experiments with Enhanced Photostability. J. Am. Chem. Soc. 131, 13610–13612. 5. Young, T. S., Ahmad, I., Yin, J. A., and Schultz, P. G. (2009) An Enhanced System for Unnatural Amino Acid Mutagenesis in E. coli. J. Mol. Biol. 15, 361–367. 6. Dirksen, A., Hackeng, T. M., and Dawson, P. E. (2006) Nucleophilic catalysis of oxime ligation. Angew. Chem. Int. Ed. Engl. 45, 7581–7584.

Chapter 2 Enzymatically Catalyzed Conjugation of a Biodegradable Polymer to Proteins and Small Molecules Using Microbial Transglutaminase Ahmed Besheer, Thomas C. Hertel, Jörg Kressler, Karsten Mäder, and Markus Pietzsch Abstract Hydroxyethyl starch (HES) is a water-soluble, biodegradable derivative of starch that is widely used in biomedicine as a plasma volume expander. Due to its favorable properties, HES is currently being investigated at the industrial and academic levels as a biodegradable polymer substitute for polyethylene glycol. To date, only chemical methods have been suggested for HESylation; unfortunately, however, these may have negative effects on protein stability. To address this issue, we have developed an enzymatic method for protein HESylation using recombinant microbial transglutaminase (rMTG). rMTG enzyme is able to catalyze the replacement of the amide ammonia at the g-position in glutamine residues (acyl donors) with a variety of primary amines (acyl acceptors), including the amino group of lysine (Lys). To convert HES into a suitable substrate for rMTG, the polymer was derivatized with either N-carbobenzyloxy glutaminyl glycine (Z-QG) or hexamethylenediamine to act as an acyl donor or acyl acceptor, respectively. Using SDS-PAGE, it was possible to show that the modified HES successfully coupled to test compounds, proving that it is accepted as a substrate by rMTG. Overall, the enzymatic approach described in this chapter provides a facile route to produce biodegradable polymer–drug and polymer–protein conjugates under relatively mild reaction conditions. Key words: Hydroxyethyl starch, Biodegradable polymer, Recombinant microbial transglutaminase, Polymer–drug conjugates, Polymer–protein conjugates

1. Introduction The surface modification of therapeutic proteins by coupling them to water-soluble polymers imparts a number of advantages, such as increased water solubility, increased circulation time, reduced levels of aggregate formation, reduced immunogenicity,

Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_2, © Springer Science+Business Media, LLC 2011

17

18

Besheer et al.

and increased stability against proteolytic digestion. The gold standard for this approach is PEGylation (coupling to poly(ethylene glycol), PEG), and many types of PEGylated proteins are already commercially available (1). However, PEG is not biodegradable, raising concerns about its fate and effects after chronic use in large doses. This has motivated academia and industry to search for biodegradable substitutes, such as albumin and hydroxyethyl starch (HES). The latter is a semisynthetic biodegradable polymer that is currently used commercially as a blood plasma volume expander. Its favorable properties, such as high water solubility, low hypersensitivity, and the possibility to tailor its molar mass and biodegradation rate, have recently attracted much attention to HES as a promising substitute for PEG. Consequently, the polymer has been investigated for the stabilization of nanoparticles (2) and for protein conjugation in order to evaluate HESylation as a potential alternative to conventional PEGylation of proteins (3). There are several chemical approaches to couple biomacromolecules to polymers; however, many of them can have detrimental effects on proteins, which are usually quite sensitive to their surrounding environment. Accordingly, enzymatic methods have been proposed and tested as gentle alternatives to chemicalbased coupling strategies. Among these, enzymatic coupling using microbial transglutaminase (MTG) has earned a significant degree of attention in the development of site-specific conjugation strategies (4). MTG catalyzes the replacement of the amide ammonia at the g-position in glutamine residues (acyl donors) with a variety of primary amines (acyl acceptors), including the e-amino group of lysine (5). Furthermore, MTG has a number of advantageous properties over eukaryotic TG – including being a calcium-independent enzyme and having fewer substrate specificity requirements (5) – and has thus found application in the food industry for crosslinking meat and fish products (5). In biomedical applications, Sato et al. provided a clear demonstration of the power of using MTG for site-specific protein–polymer conjugation when they used the enzyme to couple alkylamine derivatives of PEG selectively to a glutamine residue (Gln74) of recombinant human interleukin-2 (6). Fontana et  al. have recently reviewed the relationship between the catalytic activity of MTG and the substrate structural characteristics, and concluded that both the primary structure (viz., the presence of nearby hydrophobic residues) and tertiary structure (viz., chain flexibility) of the substrate are important factors for site specificity (7). In this chapter, we describe protocols for the modification of HES with hexamethylenediamine (HMDA) as well as with N-carbobenzyloxy glutaminyl glycine (Z-QG) to act as substrates for MTG (both as acyl acceptor and acyl donor, respectively). In addition, we present examples of the reaction of the modified

19

Enzymatically Catalyzed Conjugation of a Biodegradable Polymer to Proteins

HES polymer with model compounds to demonstrate the feasibility of this conjugation strategy. To carry out these conjugation reactions, we utilize a highly purified recombinant MTG (rMTG) carrying a polyhistidine tag (His-tag) at the C terminus, which has recently been developed in our laboratory (8, 9). The incorporation of a His-tag into rMTG facilitates the workup and separation of the products, and further provides the possibility of immobilizing the enzyme without loss of activity (10).

2. Materials 2.1. Modification of HES to Carry a Lysine-Like Residue (HES 70-Amine)

1. HES (70  kDa) (HES 70) (Serumwerke Bernburg AG, Bernburg, Germany). 2. Tosyl chloride. 3. HMDA. 4. Triethylamine. 5. 50 mM borate buffer, pH 10.

2.2. Modification of HES to Carry a Glutamine Residue (HES 70-GQ-Z)

1. HES 70 (Serumwerke Bernburg AG). 2. Dicyclohexylcarbodiimide (DCC). 3. N-Hydroxysuccinimide (NHS). 4. 4-(Dimethylamino)pyridine (DMAP). 5. N-Carbobenzyloxy glutaminyl glycine Bubendorf, Switzerland). Store at −20°C.

2.3. Reaction of HES 70-Amine with Dimethylcasein

(Bachem

AG,

1. Recombinant microbial transglutaminase (rMTG) (see Notes 1–3). 2. HES 70-Amine, as prepared in Subheading 3.2. 3. N,N-Dimethylcasein (DMC). 4. Human serum albumin (HSA), 20% (w/v). Store at 4°C. 5. Dithiothreitol (DTT). 6. 50 mM Tris–HCL buffer, pH 8. 7. Sample buffer (2×): Dissolve 1.21  g of Tris–HCl, 2.5  g of SDS, 50 mg of bromophenol blue, 10 g of glycerol, and 42 g urea in 95 ml of ultrapure water. Adjust the pH to 8.0 with HCl, and bring up to 100 ml with ultrapure water.

2.4. Reaction of HES 70-GQ-Z with Monodansyl Cadaverine

1. Recombinant microbial transglutaminase (see Notes 1–3). 2. HES 70-GQ-Z, as prepared in Subheading 3.3. 3. MDC. 4. N,N-Dimethylcasein.

20

Besheer et al.

5. HES 70 (Serumwerke Bernburg AG). 6. 50 mM Tris–HCL buffer, pH 8. 7. Sample buffer (2×) with b-mercaptoethanol: Prepare as described in Subheading 2.3, item 7, and then add 10 ml of b-mercaptoethanol/ml sample buffer before use. 2.5. SDSPolyacrylamide Gel Electrophoresis

1. Mighty Small gel electrophoresis unit (Hoefer, Inc., Holliston, MA). 2. Separating gel buffer: Add 18.18 g of Tris–HCL, 0.4 g of SDS, and 100 ml of 10% (w/v) NaN3 to 80 ml distilled water. Adjust the pH to 8.8 with 4N HCl (approximately 6 ml), and then bring up to 100 ml with water. Store at room temperature. 3. Stacking gel buffer: Add 6.06 g of Tris–HCL, 0.4 g of SDS, and 100  ml of 10% (w/v) NaN3 to 70  ml distilled water. Adjust to pH 6.8 with 4 N HCl, and then bring up to 100 ml with water. Store at room temperature. 4. 30% (w/v) acrylamide/bisacrylamide solution: Dissolve 29.1 g of acrylamide and 0.9 g of bisacrylamide in distilled water and bring up to 100 ml. CAUTION: Acrylamide is a neurotoxin and a suspected carcinogen when unpolymerized, so use extreme care and wear gloves and goggles when handling. 5. N,N,N,N ’-Tetramethylethylenediamine (TEMED). 6. 10% (w/v) Ammonium persulfate (APS): Dissolve 100 mg of APS in 1  ml of distilled water. Prepare immediately before use. 7. Water-saturated isobutanol: Shake equal volumes of water and isobutanol in a glass bottle and allow the phases to separate. Recover the organic (top) layer and store at room temperature. 8. Running buffer (10×): Mix 30.28 g of Tris–HCL, 144 g of glycine, 10 g of SDS, and 1 ml of 10% (w/v) NaN3 to distilled water and bring to 1 l. Store at room temperature. 9. Prestained molecular weight marker (Fermentas GmbH, Germany): b-Galactosidase (116.0  kDa), BSA (66.2  kDa), ovalbumin (45.0 kDa), lactate (35.0 kDa), restriction endonuclease Bsp98I (25.0 kDa), b-lactoglobulin (18.4 kDa), and lysozyme (14.4 kDa). 10. Staining solution: Dissolve one PhastGel™ Blue R (Coomassie R 350 stain) tablet (GE Healthcare) in a solution containing 50 ml of acetic acid, 100 ml of isopropanol, and 150 ml of distilled water. Accelerate the dissolution of the tablet by stirring. Filter the staining solution before use. 11. Destaining solution: Add 20 ml of isopropanol to 10 ml of acetic acid, and then bring up to 100 ml with distilled water.

Enzymatically Catalyzed Conjugation of a Biodegradable Polymer to Proteins

21

3. Methods 3.1. Modification of HES to Carry a Lysine-like Residue (HES 70-Amine) (see Note 4)

1. Dry 1 g of HES 70 (5.4 mM of the anhydroglucose units, AGU) at 105°C for 2 h. 2. Dissolve the dried HES 70 in a solution containing 10 ml of dry DMF and 1 ml of triethylamine at 60°C. 3. Dissolve tosyl chloride (0.5 g, 2.6 mM) in 1 ml of dry DMF. 4. Cool both solutions on ice to 0°C and protect from light. 5. Add the tosyl chloride solution gradually to the HES 70 solution and stir at 0°C for 1 h. 6. Precipitate the polymer solution in 100 ml of cold acetone, filter, and wash with another 100 ml of acetone. 7. Dissolve the precipitate in water and dialyze against distilled water for 3 days (6–8 kDa MWCO), then lyophilize. 8. From the prepared HES tosylate, dissolve 200 mg in 30 ml DMF/borate buffer, pH 10 (1:2, v/v). 9. Add an excess of HMDA (500 mg, 4.3 mM) dissolved in 10 ml of DMF/borate buffer, pH 10 (1:2, v/v), and stir overnight. 10. Precipitate the polymer in 200 ml of isopropanol/methanol (1:1, v/v), filter, and wash with 100 ml of the precipitating solvent. 11. Dry the precipitate at room temperature for 2 days. 12. Characterize the HES 70-amine product by 1H NMR (D2O): d = 1.27 (broad, 4H, –NH–(CH2)2–(CH2)2–(CH2)2–NH2), 1.5 (broad, 4H, –NH–CH2–CH2–(CH2)2–CH2–CH2–NH2), 5.1–5.7 (broad, 1H, HC– anomeric carbon of AGU).

3.2. Modification of HES to Carry a Glutamine Residue (HES 70-GQ-Z) (see Note 5)

1. Dissolve 185 mg (0.55 mM) of Z-QG, 114 mg (0.55 mM) of DCC, 66 mg (0.55 mM) of DMAP, and 64 mg (0.55 mM) of NHS in 2 ml of dry DMSO. Leave the mixture to react for 24 h under stirring at 400 rpm. 2. Dry 1 g of HES 70 (5.4 mM of the AGU) at 105°C for 2 h, and then dissolve the dried HES 70 in 10 ml of dry DMSO. 3. Filter the solution of activated Z-QG (Z-QG succinimidyl ester) to remove the insoluble byproduct of the reaction, dicyclohexylurea. 4. Add the filtrate from step 3 directly to the HES 70 solution (obtained from step 2) and stir for 6 h. 5. Dialyze the polymer solution against distilled water for 3 days (6–8 kDa MWCO), filter, and then lyophilize. 6. Characterize the HES 70-GQ-Z product by 1H NMR (D2O): d = 2.26 (broad, 2H, –CH2–CO–NH2), 5.03 (broad, 2H, –O–CH2–C6H5), (broad, 1H, HC– anomeric carbon of AGU), 7.31 (broad, 5H, –O–CH2–C6H5).

22

Besheer et al.

3.3. Reaction of HES 70-Amine with Dimethylcasein (see Note 6)

The reaction of HES 70-amine with DMC (step 1) can be performed simultaneously with appropriate control experiments, such as those described in steps 2 and 3 below. 1. Add 100 ml of HES 70-amine (5 mg/ml) to 100 ml of DMC (5 mg/ml), both dissolved in 0.1 M Tris–HCL buffer (pH 8). Next, add 100 ml of rMTG (15 U/ml) in glycerol to the mixture and incubate for 1 h at 37°C. 2. Positive control experiment (see Note 7): Add 200 ml of 0.2% (w/v) HSA in 0.1 M Tris–HCL buffer (pH 8) with 10 mM DTT to 100 ml of rMTG (15 U/ml) in glycerol. Incubate the reaction mixture at 37°C for 1 h. 3. Negative control experiment: Add 200 ml of DMC (5 mg/ml) in 0.1 M Tris–HCL buffer (pH 8) to 100 ml of rMTG (15 U/ ml) in glycerol. Do not add HES 70-amine. Incubate the reaction mixture at 37°C for 1 h. 4. Withdraw samples of the reaction mixture at 10, 20, 30, and 60 min, and mix with an equal volume of 2× sample buffer. Boil each sample at 99°C for 3 min. Analyze the samples by SDS-PAGE as described in Subheading 3.5 below (see Note 8). Stain the resulting gel with Coomassie Blue. Figure  1 shows an example of an SDS-PAGE gel image obtained for the coupling of HES 70-amine to DMC using rMTG.

3.4. Reaction of HES 70-GQ-Z with MDC

The reaction of HES 70-GQ-Z with MDC (step 1) can be performed simultaneously with appropriate control experiments, such as those described in steps 2, 3, and 4 below. 1. Add 25 ml of 5 mM MDC in 0.1 M acetic acid to 75 ml of a 1% (w/v) solution of HES 70-GQ-Z in 0.1  M Tris–HCL buffer (pH 8). Next, add 100 ml of rMTG (15 U/ml) in glycerol to the mixture and incubate at 37°C for 1 h. 2. Positive control experiment: Add 25  ml of 5  mM MDC in 0.1  M acetic acid to 75  ml of DMC (5  mg/ml) in 0.1  M Tris–HCL buffer (pH 8). Next, add 100 ml of rMTG (15 U/ ml) in glycerol to the mixture and incubate at 37°C for 1 h. 3. Negative control experiment 1: Add 25 ml of 5 mM MDC in 0.1 M acetic acid to 75 ml of a 1% (w/v) solution of unmodified HES 70 in 0.1 M Tris–HCL buffer (pH 8). Next, add 100 ml of rMTG (15 U/ml) in glycerol to the mixture and incubate at 37°C for 1 h. 4. Negative control experiment 2: Add 25 ml of 5 mM MDC in 0.1  M acetic acid to 75  ml of a 1% (w/v) solution of HES 70-GQ-Z in 0.1  M Tris–HCL buffer (pH 8). Next, add 100 ml of glycerol buffer (do not add rMTG) to the mixture and incubate at 37°C for 1 h.

Enzymatically Catalyzed Conjugation of a Biodegradable Polymer to Proteins

23

Fig. 1. SDS-PAGE analysis of the coupling of HES 70-amine to DMC using rMTG. Lane 1: Protein molecular weight markers, Lane 2: HES 70-amine + DMC (no rMTG), Lane 3: HES 70-amine + rMTG + DMC after 10 min, Lane 4: 20 min, Lane 5: 30 min, Lane 6: 60 min, Lane 7: DMC alone, Lane 8: DMC + rMTG after 60 min, Lane 9: HAS + DTT, and Lane 10: HSA + DTT + rMTG after 60 min. The gel was stained with Coomassie blue. Reproduced with permission from ref. 3 © 2009 John Wiley & Sons, Inc.

5. Withdraw samples (15  ml) from each experiment and mix with 2× sample buffer (15 ml). Boil at 99°C for 3 min. Analyze the samples by SDS-PAGE as described in Subheading 3.5. 6. Examine the gels under UV light (365 nm excitation filter; 520 nm emission filter). Figure 2 shows an example of a fluorescence gel image obtai­ ned for the coupling of HES 70-GQ-Z to MDC using rMTG. 3.5. SDSPolyacrylamide Gel Electrophoresis

The following procedure describes the use of a Mighty Small electrophoresis unit from Hoefer, Inc., but can be easily modified for use with other mini-vertical gel electrophoresis systems. 1. Assemble the gel sandwich stack (comprised of one notched alumina or glass plate, one rectangular glass plate and two spacers) and slide it into the casting clamp assembly. 2. With the middle screws lightly tightened, align the plates and spacers of the gel sandwich stack so that they protrude slightly (~1  mm) from the bottom of the casting clamp assembly. Next, secure the gel sandwich stack in place by tightening all the remaining screws until they are finger-tight.

24

Besheer et al.

Fig.  2. Fluorescence image of the SDS-PAGE gel obtained for the coupling of HES 70-GQ-Z to MDC using rMTG. Lane 1: HES 70-GQ-Z + rMTG + MDC, Lane 2: HES 70 + rMTG + MDC, Lane 3: HES 70-GQ-Z + MDC (no rMTG), and Lane 4: DMC + rMTG + MDC. From top to bottom, the arrows point to HES 70-GQ-Z coupled to MDC; the band of DMC coupled to MDC; and uncoupled MDC. The chemical structure of MDC is also shown below the fluorescence gel image. Reproduced with permission from ref. 3 © 2009 John Wiley & Sons, Inc.

3. Place the clamp assembly in the casting cradle, with the screws facing outward. In this position, the gel will be visible through the rectangular glass plate. Ensure that there is a good seal between the bottom of the gel sandwich stack and the rubber gasket in the casting cradle.

Enzymatically Catalyzed Conjugation of a Biodegradable Polymer to Proteins

25

4. Prepare the separating gel in the following order (stir the mixture before and after adding APS to ensure homogeneous polymerization): Distilled water (3.3 ml), separating gel buffer (2.5  ml), acrylamide/bisacrylamide (4.2  ml), TEMED (10 ml), and APS (20 ml). 5. Pour the separating gel into the sandwich stack. Fill the sandwich stack until the gel solution reaches a few centimeters below the top of the rectangular glass plate so that there is sufficient room left for pouring the stacking gel (step 9). 6. Overlay the separating gel with 500  ml of isopropanol and allow the gel to polymerize for 30 min. 7. Remove the isopropanol by simply inverting the whole gel casting assembly and discarding the drained liquid into an organic solvent waste container. 8. Prepare the stacking gel in the following order (stir the mixture before and after adding APS to ensure homogenous polymerization): Distilled water (3  ml), stacking gel buffer (1.25  ml), acrylamide/bisacrylamide (0.75  ml), TEMED (10 ml), and APS (8 ml). 9. Pour the stacking gel onto the separating gel until the gel solution reaches the top of the glass plate. Insert the comb and allow the gel to polymerize. 10. Prepare the running buffer by diluting 30 ml of 10× concentrated running buffer stock solution with 270 ml of distilled water. 11. Transfer the polymerized gel sandwich stack to the electrophoresis unit and secure it in place. 12. Connect the coolant ports of the electrophoresis unit to a circulating (cold) water bath. 13. To aid in loading the samples into the gel, wet the transparent well-locating decal and apply it to the front of the glass plate so that the appropriate edge outlines the sample wells. 14. Fill the upper buffer chamber with running buffer and then remove the comb. 15. Load the samples into the wells of the gel (load 10 ml of the samples and 5 ml of the protein molecular weight markers). 16. Fill the lower buffer chamber with running buffer and install the safety lid onto the electrophoresis unit. 17. Connect the electrophoresis unit to a power supply and run the gel at 40 mA (constant current mode) for ~6 min through the stacking gel, and then run the gel at 30 mA for ~40 min through the separating gel. (When running two gels, run at 80  mA for 6  min through the stacking gel and 60  mA for 40 min through the separating gel.)

26

Besheer et al.

18. After the bromophenol blue tracking dye has reached the end of the gel (or has passed completely through the gel), turn off the power and disconnect the electrophoresis unit from the power supply. Also, stop the circulating water bath. 19. To take the gel out of the sandwich stack, remove the spacers first and then carefully pry the aluminum backing plate with a spatula to strip the gel from the glass plate. 20. Briefly rinse the gel with distilled water. 21. To stain the protein bands, gently agitate the gel in staining solution for ~1 h. 22. Rinse the gel briefly with destaining solution, and then agitate the gel in destaining solution for 30 min. Replace with fresh destaining solution and agitate the gel for an additional 30 min (at this point, the gel background should not contain any blue color). 23. Analyze the stained gel using a gel imaging system.

4. Notes 1. The recombinant biocatalyst (inactive pro-TG) can be produced in soluble form using Escherichia coli and IPTG as an inductor as described in detail in ref. 9. A very important point to note is the temperature shift prior to induction (from 37°C to 24°C) to prevent the formation of inclusion bodies. This step reduces the transcription and translation velocity, delivering the pro-enzyme in limited amounts and enabling the protein to fold properly. Alternatively, the autoinduction medium (9) can be used at an incubation temperature of 28°C. 2. Inactive pro-TG enzyme can be purified using metal affinity chromatography and is activated prior to use by removal of the pro-sequence using a suitable protease, such as dispase (8). The endogenous protease from Streptomyces mobaraensis can also be used instead; however, this enzyme is not commercially available and must be prepared in the laboratory. 3. In principle, commercially available transglutaminase can be used instead of the recombinant enzyme. However, commercial preparations of transglutaminase can in some cases be relatively crude and may contain proteases (which may not be visible on a Coomassie-stained SDS-PAGE gel). Such proteases can cause unwanted degradation of the protein substrate. 4. MTG is known to display a high degree of selectivity for glutamine (Gln) as the acyl donor, while the acyl acceptor can either be a lysine (Lys) residue or simply a primary amine

Enzymatically Catalyzed Conjugation of a Biodegradable Polymer to Proteins

27

group attached to an alkyl chain having at least four carbon atoms. Accordingly, we used HMDA for the modification of HES, as well as MDC as a model acyl acceptor, where both have primary amine groups attached to alkyl chains with six and five carbon atoms, respectively. 5. Due to the fact that DCC can lead to the oxidation of alcohols in DMSO, the coupling of Z-QG to HES 70 is carried out in two steps: First, the activation of Z-QG using DCC and NHS is performed. Following this, the coupling of HES 70 to the activated peptide is achieved. 6. DMC is derived from casein by methylation (i.e., blocking) of all the lysine amino groups. Consequently, DMC can act as an acyl donor (i.e., presenting only glutamine residues) and is a useful compound for performing test conjugation reactions with acyl acceptors. 7. HSA functions as a positive control, where it is used to show that the purified rMTG has retained its bioactivity. In the native form, HSA does not act as a substrate for MTG and must first be denatured by reduction using DTT. 8. Since both HES 70-amine and DMC are polyvalent compounds, their crosslinking with rMTG can lead to the formation of very large aggregates, which can be seen as bands at the top of the stacking gel, as shown in Fig. 1. References 1. Duncan, R. (2003) The dawning era of polymer therapeutics. Nature Rev. Drug Discovery 2, 347–360. 2. Besheer, A., Vogel, J., Glanz, D., Kressler, J., Groth, T., Mäder, K. (2009) Characterization of PLGA nanospheres stabilized with amphiphilic polymers: Hydrophobically modified hydroxyethyl starch vs pluronics. Mol. Pharm. 6, 407–415. 3. Besheer, A., Hertel, T.C., Kressler, J., Mäder, K., Pietzsch, M. (2009) Enzymatically catalyzed HES conjugation using microbial transglutaminase: Proof of feasibility. J. Pharm. Sci. 98, 4420–4428. 4. Sato, H. (2002) Enzymatic procedure for sitespecific pegylation of proteins. Adv. Drug. Delivery Rev. 54, 487–504. 5. Yokoyama, K., Nio, N., Kikuchi, Y. (2004) Properties and applications of microbial transglutaminase. Applied Microbiol. Biotech. 64, 447–454. 6. Sato, H., Hayashi, E., Yamada, N., Yatagai, M., Takahara, Y. (2001) Further studies on the

site-specific protein modification by microbial transglutaminase. Bioconj. Chem. 12, 701–710. 7. Fontana, A., Spolaore, B., Mero, A., Veronese, F.M. (2008) Site-specific modification and PEGylation of pharmaceutical proteins mediated by transglutaminase. Adv. Drug Delivery Rev. 60, 13–28. 8. Marx, C. K., Hertel, T. C., Pietzsch, M. (2008) Purification and activation of a recombinant histidine-tagged protransglutaminase after soluble expression in E. coli and partial characterization of the active enzyme. Enz. Microbial. Tech. 42, 568–575. 9. Marx, C. K., Hertel, T. C., Pietzsch, M. (2007) Soluble expression of a pro-transglutaminase from Streptomyces mobaraensis in Escherichia coli. Enz. Microbial. Tech. 40, 1543–1550. 10. Cass, A. E. G., Zhang, J. K. (2001) A Study of His-tagged alkaline phosphatase immobilization on a nanoporous nickel-titanium dioxide film. Anal. Biochem. 292, 307–310.

Chapter 3 Synthesis of Drug/Dye-Incorporated Polymer–Protein Hybrids Sukanta Dolai, Wei Shi, Bikash Mondal, and Krishnaswami Raja Abstract We present here a general methodology for significantly increasing the number of dye/drug molecules that can be attached per protein molecule. As a demonstration of this approach, poly(acrylic acid) (PAA)based near-infrared fluorescence (NIRF) dye- and glucose-incorporated novel copolymers were synthesized, which were further employed for bioconjugation to avidin and bovine serum albumin (BSA). In this method, azide-terminated poly(tert-butyl acrylate) was synthesized via atom transfer radical polymerization (ATRP). Subsequent deprotection was performed to yield poly(acrylic acid) (PAA) possessing a reactive chain-end. A one-pot sequential amidation of the PAA with the amine derivatives of a nearinfrared fluorescent dye (ADS832WS) and glucose produced NIRF dye-incorporated water-soluble copolymers. End-group modifications were performed to produce alkyne/biotin-terminated copo­lymers, which were further employed to generate dye-incorporated polymer–protein hybrids via the biotin– avidin interaction with avidin or by “click” bioconjugation with azide-modified BSA. Key words: Protein–polymer hybrids, Copolymers, Near infrared fluorescence dye, “Click” bioconjugation

1. Introduction Polymer–protein hybrids are a newly emerging class of bioconjugates with several applications in biotechnology, biopharmaceutical chemistry, and other life science areas (1–14). Drug- and dyelabeled conjugates are a multibillion dollar industry; however, by using the current technologies, the number of copies of cytotoxic drugs/dyes that can be chemically conjugated to a single protein is very limited. For example, in the case of Mylotarg® (a calicheamicin–antibody conjugate), the drug to antibody ratio is 2:1. A fundamental limitation in the detection and therapeutic efficiency of imaging agent/drug-labeled proteins such as antibodies arises Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_3, © Springer Science+Business Media, LLC 2011

29

30

Dolai et al.

from the fact that only a limited number of these molecules can be attached per protein molecule, and extensive modification of proteins with several copies of a drug/dye would cause deactivation of the protein’s functional properties. On the other hand, the synthesis of copolymer–protein hybrid materials, for labeling applications, has so far been restricted to only a few types of acrylate/methacrylate monomers due to differences in solubility between monomers; for instance, the preparation of a copolymer, containing a monosaccharide-derived acrylate (hydrophilic) and taxol acrylate (hydrophobic), would be very challenging. The widely different reactivity of the monomers and the lack of reactivity of many biologically relevant acrylates are other factors that are responsible for the lack of diversity of copolymer–protein hybrids (e.g., the acrylate of the anticancer and antiAlzheimer’s drug candidate curcumin can be synthesized but cannot be polymerized via free radical polymerization methods because the molecule is a radical scavenger (15)). We have recently developed a general methodology for significantly increasing the number of dye/drug molecules that can be attached per protein molecule (14). Herein, we present detailed practical procedures for (1) the synthesis of well-defined living copolymers containing reactive chain-end and functional sidechain pendant groups in which the chain end and side chains possess orthogonal reactivity, (2) the attachment of a number of dye and glucosamine molecules to the functional polymer side-chains, and (3) the attachment of the polymers (via the reactive polymer chain end) with proteins to produce the final bioconjugates.

2. Materials 2.1. Synthesis of Biotin-Terminated Poly(NIRF)–Poly(Glu) Polymer

1. Biotin-terminated poly(acrylic acid) (PAA) polymer. This polymer is synthesized according to the procedure previously reported in ref. 14. 2. Near infrared absorption dye (ADS832WS) (American Dye Source, Inc., Baie D’Urfé, Quebec). 3. 1-Ethyl-3-(3-dimethylaminopropyl)carbodiimide hydrochloride (EDC⋅HCl). 4. N-Hydroxybenzotriazole (HOBt). 5. N,N-Dimethylformamide (DMF). 6. Triethylamine. 7. d(+)-Glucosamine. 8. Milli-Q water (Millipore). 9. Dialysis tubing (3.5 kDa MWCO). 10. Sephadex™ LH-20 gel filtration chromatography medium (GE Healthcare).

Synthesis of Drug/Dye-Incorporated Polymer–Protein Hybrids

2.2. Synthesis of Avidin–Poly (NIRF)–Poly(Glu) Polymer Conjugate

1. Biotin-terminated Subheading 3.1).

poly(NIRF)–poly(Glu)

polymer

31

(see

2. Avidin. 3. 0.1 M Phosphate-buffered saline (PBS), pH 7.4. 4. Dialysis tubing (50 kDa MWCO). 5. Milli-Q water (Millipore).

2.3. Size-Exclusion Chromatography Analysis of Avidin– Polymer Conjugate

1. Fast protein liquid chromatography (FPLC) system (AKTA™ Explorer system from GE Healthcare) equipped with a multiple wavelength detector. 2. HiPrep™ Sephacryl™ S-200 HR 26/10 gel filtration column (GE Healthcare). 3. 0.1 M PBS, pH 7.4.

2.4. SDS-PAGE Analysis of Conjugate

1. PAGE gel 4–20% SDS Cassette Gel precast gels, 12-well (Fisher Scientific). 2. PAGEgel DTT reducer, 10× (Fisher Scientific). 3. PAGEgel LDS Sample buffer, 4× (Fisher Scientific). 4. PAGEgel SDS Standard Run buffer, 20× (Fisher Scientific). 5. PAGEgel Two-Color SDS Marker (Orange/Blue) (Fisher Scientific). Ready-to-use in a 1× LDS sample buffer. 6. GelCode Blue coomassie gel stain reagent (Fisher Scientific). 7. FisherBiotech FB1000 electrophoresis power supply. 8. VWR™ rocking platform. 9. Odyssey® Infrared Imaging System (LI-COR Biosciences, Lincoln, NE).

2.5. Synthesis of Azide-Terminated Poly(NIRF)–Poly(Glu) Polymer

1. Azide-terminated poly(acrylic acid) (PAA) polymer. This polymer is synthesized according to the procedure previously reported in ref. 14 (see Note 1).

2.6. Synthesis of Alkyne-Terminated Poly(NIRF)–Poly(Glu) Polymer

1. Azide-terminated Subheading 3.5).

2. All other required materials are the same as those listed in Subheading 2.1, items 2–10. poly(NIRF)–poly(Glu)

polymer

2. 3-Prop-2-ynoxyprop-1-yne (dipropargyl ether). 3. Copper sulfate pentahydrate (CuSO4⋅5H2O). 4. Sodium ascorbate. 5. Tert-Butanol (t-BuOH). 6. Tetrahydrofuran (THF). 7. Milli-Q water (Millipore).

(see

32

Dolai et al.

8. Dialysis tubing (10 kDa MWCO). 9. Sephadex™ LH-20 gel filtration chromatography medium (GE Healthcare). 2.7. AzideModification of BSA (Protein Model)

1. Bovine serum albumin (BSA). 2. 2,5-Dioxopyrrolidin-1-yl-5-azidopentanoate (NHS-azide). This reagent can be synthesized according to the procedure previously reported in ref. 14. 3. Dimethyl sulfoxide (DMSO). 4. 0.1 M Tris buffer, pH 8.0. 5. Dialysis tubing (10 kDa MWCO).

2.8. “Click” Bioconjugation of BSA–Azide with Alkyne-Terminated Poly(NIRF)–Poly(Glu) Polymer

1. Alkyne-terminated Subheading 3.6).

poly(NIRF)–poly(Glu)

polymer

(see

2. Azide-modified BSA (see Subheading 3.7). 3. Tris(2-carboxyethyl)phosphine (TCEP). 4. Tris((1-phenyl-1H-1,2,3-triazol-4-yl)methyl)amine (TPTA). TPTA is a polytriazolylamine ligand which stabilizes Cu(I) toward disproportionation and oxidation, thus enhancing its catalytic effect in alkyne-azide cycloaddition (“click”) reactions. 5. Copper sulfate pentahydrate (CuSO4 ⋅ 5H2O). 6. 0.1 M Tris buffer, pH 8.0. 7. Dimethylformamide (DMF). 8. 0.1 M PBS, pH 7.4. 9. Dialysis tubing (50 kDa MWCO).

2.9. Size-Exclusion Chromatography Analysis of BSA– Poly(NIRF)–Poly(Glu) Polymer Conjugate

1. All the required materials are the same as those listed in Subheading 2.3.

2.10. SDS-PAGE Analysis of BSA– Polymer Conjugate

1. All the required materials are the same as those listed in Subheading 2.4, items 1–9.

3. Methods Using the procedures described below, a one-pot sequential amidation of the PAA with the amine derivative of a nearinfrared fluorescent dye (NIRF) (ADS832WS) and glucose produced NIRF dye-incorporated water-soluble copolymers.

Synthesis of Drug/Dye-Incorporated Polymer–Protein Hybrids

33

End-group modifications were also performed to produce alkyne/ biotin-terminated copolymers, which were further employed to generate dye-incorporated polymer–protein hybrids via the biotin–avidin interaction with avidin or by “click” bioconjugation (16) with azide-modified BSA. With these general methods, two fundamental limitations in the synthesis of bioconjugates are overcome: (1) the basic restriction in the diversity of copolymers that can be synthesized for producing bioconjugates and (2) the limitation in the number of dyes/drug molecules that can be attached per protein molecule. 3.1. Synthesis of Biotin-Terminated Poly(NIRF)–Poly(Glu) Polymer

1. Dissolve the biotin-terminated poly(acrylic acid) (PAA) polymer (76  mg, 1.055  mmol), near-infrared absorption dye (ADS832WS) (498 mg, 0.531 mmol), EDC ∙ HCl (213 mg, 1.11  mmol), and HOBt (153  mg, 1.148  mmol) in DMF (1.5 mL) in a round-bottom flask. Place a magnetic stir bar inside the flask, and then cap it with a rubber septum. 2. Add triethylamine (0.03 mL, 0.17 mmol) via a syringe. 3. Purge the flask with N2 gas and stir the reaction mixture (using a magnetic stirrer) for 48 h at room temperature under a slow continuous flow of N2 gas. 4. After stirring for 48 h, prepare a mixture of d(+)-glucosamine (230 mg, 1.067 mmol) and EDC ∙ HCl (200 mg, 1.043 mmol) in a mixture of Milli-Q water (2 mL) and DMF (1 mL), and then add this solution to the previous reaction mixture via a syringe (see Note 2). 5. Stir the mixture for another 72 h at room temperature under a continuous flow of N2 gas. 6. After stopping the reaction, transfer the mixture into a dialysis bag (3.5  kDa MWCO) and dialyze extensively against deionized water for 24 h at room temperature (see Note 3). 7. Filter the dialyzed reaction mixture via gravity filtration and lyophilize to yield a green-brown cotton-like crude product. 8. Prepare a size-exclusion chromatography column by suspending the Sephadex™ LH-20 gel filtration medium in Milli-Q water and packing it inside a clean chromatographic column. 9. Dissolve the crude product obtained in step 7 in a minimum volume of Milli-Q water and load it onto the top of the Sephadex™ column. Elute the sample from the column with water and collect the eluted fractions. Combine the fractions containing the biotin-terminated poly(NIRF)–poly(Glu) polymer and lyophilize. 10. Characterize the purified product using nuclear magnetic resonance (NMR) spectroscopy, Fourier transform infrared (FT-IR) spectroscopy, and size-exclusion chromatography (SEC) techniques.

34

Dolai et al.

3.2. Synthesis of Avidin–Poly (NIRF)–Poly(Glu) Polymer Conjugate

1. Dissolve avidin (1  mg) in 0.25  mL of PBS buffer, pH 7.4. Also separately dissolve biotin-terminated poly(NIRF)– poly(Glu) polymer (obtained in Subheading 3.1) (5.5 mg) in 0.65  mL PBS buffer (pH 7.4), and then slowly add this solution to the avidin solution. 2. Incubate the resulting mixture for 2 h at room temperature. 3. After incubation, transfer the mixture into a dialysis bag (50 kDa MWCO) and dialyze extensively against deionized water for 24 h at 4°C (see Note 3). 4. Analyze the protein–polymer hybrid by size-exclusion FPLC (Subheading 3.3) and SDS-PAGE (Subheading 3.4).

3.3. Size-Exclusion Chromatography Analysis of the Hybrid

Perform the size-exclusion FPLC analysis procedure using the following samples: avidin protein (control), biotin-terminated poly(NIRF)–poly(Glu) polymer (control, obtained from Subheading 3.1), and the avidin–poly(NIRF)–poly(Glu) polymer conjugate (obtained from Subheading 3.2). 1. Attach the HiPrep™ Sephacryl™ S200 HR 26/10 gel filtration column to a FPLC system equipped with a sample injection loop of 1 mL, and set the detector to monitor at 260, 280, and 700 nm wavelengths. 2. Prepare a sufficient volume of running buffer (~4 L of PBS, pH 7.4) for the entire run and place it in the FPLC buffer reservoir. 3. Wash the column with PBS, pH 7.4 for at least two column volumes (i.e., about 640 mL), using a flow rate of 1 mL/min. 4. Set up the following chromatography method file: Flow rate = 1  mL/min, total run = 1.5 column volumes; perform column equilibration for 0.2 column volumes before sample injection, sample injection volume = 1 mL. 5. Start the chromatography run, and load 1 mL of the sample (~5  mg/mL) into the injector module via a syringe; make sure there are no air bubbles. 6. Compare the results of the FPLC analysis for the different samples. Representative FPLC chromatogram profiles for avidin, biotin-terminated poly(NIRF)–poly(Glu) polymer, and the avidin– poly(NIRF)–poly(Glu) polymer conjugate are shown in Fig. 1.

3.4. SDS-PAGE Analysis of Avidin– Polymer Conjugate

Perform the SDS-PAGE analysis procedure using the following samples: Avidin protein (control), biotin-terminated poly(NIRF)– poly(Glu) polymer (control, obtained from Subheading 3.1), and the avidin–poly(NIRF)–poly(Glu) polymer conjugate (obtained from Subheading 3.2). The instructions described below assume

Synthesis of Drug/Dye-Incorporated Polymer–Protein Hybrids

35

Fig.  1. Size-exclusion FPLC analysis of avidin, biotin-terminated poly(NIRF)–poly(Glu) copolymer, and the avidin–polymer conjugate. The avidin–polymer conjugate eluted earlier (due to its higher molecular weight) in comparison to the avidin control sample and the polymer control sample. Reproduced with permission from ref. 14 © 2007 American Chemical Society.

the use of Fisher Scientific PAGEgel SDS electrophoresis reagents, but are easily adaptable for use with other standard SDS-PAGE gel systems. 1. Load a precast SDS cassette gel into the electrophoresis unit, and insert a dummy plate on the other side of the apparatus. Ensure that the cassette gel and the dummy plate are mounted tightly. 2. Prepare 1× running buffer by taking 60  mL of the SDS Standard Run buffer 20× stock solution and mixing it with 1,140 mL of Milli-Q water. Add the 1× SDS Standard Run buffer to the upper (inner) chamber of the electrophoresis apparatus, and confirm that there is no leakage. Next, fill the chamber with enough running buffer to cover the cassette gel completely. 3. Wash each well in the cassette gel with running buffer using a sample loading tip and a pipettor. 4. Prepare the avidin (control) sample: Add 30  mL of avidin (~3 mg/mL), 4 mL of DTT reducer, and 10 mL of LDS sample buffer to an Eppendorf tube and mix by vortexing. Place the sample in a float and boil it in a water bath for 4  min. Remove the sample and place it aside on ice.

36

Dolai et al.

5. Prepare the polymer (control) sample: Add 30 mL of biotinterminated poly(NIRF)–poly(Glu) polymer (~3  mg/mL), 4 mL of DTT reducer, and 10 mL of LDS sample buffer to an Eppendorf tube and mix by vortexing. Place the sample aside on ice. 6. Prepare the avidin–polymer conjugate sample: Add 30 mL of avidin–poly(NIRF)–poly(Glu) polymer conjugate (~3  mg/ mL), 4 mL of DTT reducer, and 10 mL of LDS sample buffer to an Eppendorf tube and mix by vortexing. Place the sample in a float and boil it in a water bath for 4 min. Remove the sample and place it aside on ice. 7. Prepare a blank sample: Add 30 mL of PBS, pH 7.4 and 10 mL of LDS sample buffer to an Eppendorf tube and mix by vortexing. Place the sample aside on ice. 8. Carefully load the protein/polymer samples (i.e., avidin control, polymer control, and polymer conjugate), protein molecular weight markers, and the blank solution into the wells in the cassette gel by placing the tip of the loading pipette at the bottom of each well and slowly drawing the tip upward as you fill up the wells. Load the blank solution between each of the sample wells. 9. Fill the lower (outer) chamber of the electrophoresis apparatus with running buffer up to the halfway mark on the cassette gel. 10. Place the top safety cover onto the electrophoresis apparatus and connect it to a power source. Turn the power source on, and set the voltage to 90 V. Start the electrophoresis run; the tracking dye in the samples should begin to travel downward. Stop the power when you see the tracking dye reaches the bottom of the gel cassette. 11. Disconnect the power supply and remove the top safety cover of the electrophoresis apparatus. Slowly remove the cassette gel. Remove the outer plastic cover of the cassette gel using a spatula, and carefully place the gel inside a rectangular container. Rinse the gel carefully with Milli-Q water. 12. After rinsing, place the gel onto the scanner bed of an infrared imaging system. Put a small amount of water on top of the gel and on the scanner plate so that the gel does not dry out. Scan the gel at medium resolution to acquire an infrared fluorescence image (EX l = 800  nm for NIRF dye ADS832WS). 13. Place the gel back into the container after scanning and add a sufficient volume of GelCode Blue coomassie stain reagent to completely cover the gel. Place the gel onto a rocking platform and agitate it for 30 min at room temperature.

Synthesis of Drug/Dye-Incorporated Polymer–Protein Hybrids

37

Fig.  2. SDS-PAGE analysis of avidin (lane a) and avidin–poly(NIRF)–poly(Glu) polymer conjugate (lane b). The formation of the conjugate is indicated by the higher molecular weight bands present in the SDS-PAGEgel; the conjugate band that glowed when imaged with an infrared scanner (left ) is also visible following Coomassie staining (right ), thus confirming the presence of both polymer and protein at the same position. Reproduced with permission from ref. 14 © 2007 American Chemical Society.

14. Discard the liquid stain and wash the gel three times with Milli-Q water. 15. Compare the image of the coomassie-stained gel with the infrared fluorescence image of the scanned gel to confirm the formation of the avidin–polymer conjugate (Fig. 2). 3.5. Synthesis of Azide-Terminated Poly(NIRF)–Poly(Glu) Polymer

1. Dissolve the azide-terminated poly(acrylic acid) (PAA) polymer (72  mg, 1.003  mmol), near-infrared absorption dye (ADS832WS) (100 mg, 0.531 mmol), EDC∙HCl (213 mg, 1.11  mmol), and HOBt (153  mg, 1.148  mmol) in DMF (1.5 mL) in a round-bottom flask. Place a magnetic stir bar inside the flask, and then cap it with a rubber septum. 2. Add Triethylamine (0.03 mL, 0.17 mmol) via a syringe. 3. Purge the flask with N2 gas and stir the reaction mixture (using a magnetic stirrer) for 48 h at room temperature under a slow continuous flow of N2 gas.

38

Dolai et al.

4. After stirring for 48 h, prepare a mixture of d(+)-glucosamine (190 mg, 0.9 mmol) and EDC ∙ HCl (200 mg, 1.043 mmol) in a mixture of Milli-Q water (2 mL) and DMF (1 mL), and then add this solution to the previous reaction mixture via a syringe (see Note 2). 5. Stir the mixture for another 72 h at room temperature under a continuous flow of N2 gas. 6. After stopping the reaction, transfer the mixture into a dialysis bag (10  kDa MWCO) and dialyze extensively against deionized water for 24 h at room temperature (see Note 3). 7. Filter the dialyzed reaction mixture via gravity filtration and lyophilize to yield a green-brown cotton-like crude product. 8. Proceed to purify the product by gel filtration chromatography as described in Subheading 3.1, steps 8 and 9. 9. Characterize the purified product using NMR, FT-IR, and SEC techniques. 3.6. Synthesis of Alkyne-Terminated Poly(NIRF)–Poly(Glu) Polymer

1. Dissolve azide-terminated poly (NIRF)–(Glu) polymer (obtained in Subheading 3.1) (60 mg, 3 mmol) in 3 mL of a mixture of t-BuOH, THF, and Milli-Q water (1:1:1, v/v/v) in a round-bottom flask. Place a magnetic stir bar inside the flask, and then cap it with a rubber septum. 2. Add 10  mL of propargyl ether (9.14  mg, 31.5  eq) via a syringe. 3. Add sodium ascorbate (4  mg, 0.02  mmol) to the reaction vessel while stirring (using a magnetic stirrer). Finally, add CuSO4 ⋅ 5H2O (5 mg, 0.02 mmol) to the reaction mixture. 4. Purge the reaction vessel with N2 gas and continue to stir the mixture for 24 h at room temperature. 5. After stopping the reaction, transfer the mixture into a dialysis bag (10  kDa MWCO) and dialyze extensively against deionized water for 24 h at room temperature (see Note 3). 6. Proceed to purify the product by gel filtration chromatography as described in Subheading 3.1, steps 8 and 9. 7. Characterize the purified product using NMR, FT-IR, and SEC techniques.

3.7. Modification of the Amine Surface Groups of BSA with NHS-Azide

1. Dissolve 5 mg of BSA in 1 mL of PBS buffer, pH 7.4 in a small vial and gently agitate on a vortex mixer. Do not excessively shake the mixture in order to avoid the formation of foam. 2. Next, measure 5 mg of NHS-azide (10 eq to each modifiable lysine group on BSA) in 100 mL of DMSO. 3. Add the NHS-azide dropwise to the BSA solution while agitating on a vortex. Add the NHS-azide slowly enough so that nothing precipitates out. Mix the components thoroughly and put the

Synthesis of Drug/Dye-Incorporated Polymer–Protein Hybrids

39

vial containing the mixture on a rocking platform and let it shake gently for 3 h at room temperature. 4. Transfer the mixture to a 10-kDa MWCO dialysis membrane and dialyze against 0.1  M Tris buffer, pH 8.0 for 48  h to remove excess azide linker (see Note 3). Transfer the dialyzed solution to a container and store it at 4°C. 3.8. Conjugation of the AlkyneTerminated Poly(NIRF)–Poly(Glu) Polymer with AzideModified BSA

1. Add 20  mg of alkyne-terminated poly(NIRF)–poly(Glu) (obtained from Subheading 3.6) to a vial containing 1.8 mL of azide-modified BSA (obtained from Subheading 3.7). The starting volume of this reaction mixture should be about 1.8 mL. 2. Calculate the amount of TCEP, TPTA, and copper sulfate required in order to make the concentration of these reagents in the final reaction mixture (about 2  mL) to be 4, 4, and 2 mM, respectively. 3. Add the required amount of TCEP to the vial containing the alkyne-terminated poly(NIRF)–poly(Glu) polymer and azidemodified BSA. Mix the resulting solution over a vortex. 4. Dissolve the required amount of TPTA in 200  mL of DMF. The amount of DMF should not exceed more than 10% (v/v) of the total reaction volume (if the DMF amount exceeds 10%, there is a chance that protein precipitation will occur.) Slowly add the TPTA solution to the reaction mixture dropwise while agitating on a vortex. Ensure that no protein precipitates out. If the solution appears cloudy, stop adding TPTA and wait until the solution clears before continuing to add the reagent (see Note 4). 5. Add the required amount of copper sulfate to the reaction mixture and agitate on a vortex. Note that it is necessary to add the TPTA reagent to the reaction mixture before the addition of copper sulfate. 6. Incubate the reaction mixture on an agitator at 4°C for 16 h. Next, transfer the reaction mixture to a 50-kDa MWCO dialysis membrane and dialyze against PBS buffer, pH 7.4 for 48 h to remove excess reagents and byproducts (see Note 3). 7. Characterize the protein–polymer hybrid by size-exclusion FPLC (Subheading 3.9) and SDS-PAGE (Subheading 3.10).

3.9. Size-Exclusion Chromatography Analysis of the BSA–Polymer Conjugate

Perform the size-exclusion FPLC analysis procedure, as described in Subheading 3.3, using the following samples: azide-modified BSA protein (control), alkyne-terminated poly(NIRF)–poly(Glu) polymer (control, obtained from Subheading 3.6), and the BSA– poly(NIRF)–poly(Glu) polymer conjugate (obtained from Subheading 3.8). Compare the results of the FPLC analysis for the different samples. Representative FPLC chromatogram profiles for azide-modified BSA, alkyne-terminated poly(NIRF)–poly(Glu) polymer, and the BSA–poly(NIRF)–poly(Glu) polymer conjugate are shown in Fig. 3.

40

Dolai et al.

Fig. 3. Size-exclusion FPLC analysis of azide-modified BSA (dashed line), alkyne-terminated poly(NIRF)–poly(Glu) polymer (dotted line), and the BSA–poly(NIRF)–poly(Glu) polymer “click” conjugate (solid line). The conjugate eluted earlier than both the azide-modified BSA control sample and the polymer control sample. (The unmodified BSA eluted at the same volume as the azide-modified BSA; hence it was omitted from the chromatogram.) Reproduced with permission from ref. 14 © 2007 American Chemical Society.

Fig. 4. SDS-PAGE analysis of the BSA–poly(NIRF)–poly(Glu) polymer conjugate (lanes 4 and 8 ), a mixture of azide-modified BSA and polymer without “click” catalyst reagents (lane 6 ) and unmodified BSA (lane 10 ). The conjugate bands in lanes 4 and 8 that glowed when imaged with an infrared scanner (right ) are also visible following Coomassie staining (left ), thus confirming the presence of both polymer and protein at the same positions. Conversely, the control samples in lanes 6 and 10 do not glow at all when imaged with an infrared scanner. To confirm that the copolymer is indeed chemically bonded to the protein, we incubated the mixture of alkyne-terminated polymer and azide-modified BSA in the same ratio but without any “click” catalyst reagents, and then dialyzed the mixture with a 50-kDa MWCO membrane; in this case, no higher molecular weight conjugate bands are observed (lane 6 ). Reproduced with permission from ref. 14 © 2007 American Chemical Society.

Synthesis of Drug/Dye-Incorporated Polymer–Protein Hybrids

3.10. SDS-PAGE Analysis of the Conjugate

41

Perform the SDS-PAGE analysis procedure as described in Subheading 3.4 using the following samples: unmodified/azidemodified BSA protein (controls), alkyne-terminated poly(NIRF)– poly(Glu) polymer (control, obtained from Subheading 3.6), and the BSA–poly(NIRF)–poly(Glu) polymer conjugate (obtained from Subheading  3.8). Compare the results of the SDS-PAGE analysis for the different samples. Representative coomassiestained gel images and infrared fluorescence gel images for unmodified/azide-modified BSA, alkyne-terminated poly(NIRF)– poly(Glu) polymer, and the BSA–poly(NIRF)–poly(Glu) polymer conjugate are shown in Fig. 4.

4. Notes 1. Azide-terminated poly (tert-butyl acrylate) is synthesized via atom transfer radical polymerization (ATRP). Subsequent deprotection is performed to yield poly(acrylic acid) (PAA) possessing a reactive chain-end. When synthesizing the poly(tert-butyl acrylate) via ATRP, make sure that the reaction vessel is oxygen-free, as singlet oxygen species act as radical scavengers and will stop the polymerization process. Also, do not perform the polymerization reaction at temperatures higher than 70°C with azide-terminated initiator, as higher temperatures could destroy the azide functionality. After the purification of the polymer, confirm the presence of the azide group in the FT-IR spectrum around 2,100 cm−1. 2. When performing the amidation reaction between the biotin-terminated poly(acrylic acid) (PAA) polymer with NIRF-NH2 and glucosamine using EDC⋅HCl and HOBt, remember to add the NIRF-NH2 first and let it react for 48 h; then add the glucosamine to the same reaction vessel. It is also possible to add both reagents at the same time; however, due to their reactivity differences, the glucosamine will react much faster than the NIRF-NH2. Hence, the final amount of NIRF-dye loaded onto the polymer will be much lower than expected. 3. When performing dialysis, replace the dialysis solution every 2–3  h. Typically, five to six buffer exchanges using a total volume of 500 mL of buffer are needed to effectively dialyze the samples. 4. While performing the “click” bioconjugation reaction using the copper sulfate, sodium ascorbate, and TPTA reagent, remember to add sodium ascorbate or TPTA prior to the addition of the copper sulfate.

42

Dolai et al.

Acknowledgments We sincerely thank Dr. K. B. Sharpless for providing us with the TPTA “click” ligand. The National Science Foundation (CHE0723028) and PSC-CUNY provided financial support for this work. References 1. Hermenson, G. T. (2008) Bioconjugate Techniques, 2nd ed., Academic Press, San Diego, CA. 2. Wu, A. M., and Senter, P. D. (2005) Arming antibodies: prospects and challenges for immunoconjugates. Nat. Biotechnol. 23, 1137–1146. 3. Raja, K., McDonald, R., Tuck, S., Rodriguez, R., Milley, B., and Traquina, P. (2007) Onepot synthesis, purification, and formulation of bionanoparticle-CpG oligodeoxynucleotide hepatitis B surface antigen conjugate vaccine via tangential flow filtration. Bioconjugate Chem. 18, 285–288. 4. Raja, K. S., Wang, Q., Gonzalez, M. J., Manchester, M., Johnson, J. E., and Finn, M. G. (2003) Hybrid virus-polymer materials. 1. Synthesis and properties of PEG-decorated cowpea mosaic virus. Biomacromolecules 4, 472–476. 5. Kulkarni, S., Schilli, C., Grin, B., Mueller, A. H. E., Hoffman, A. S., and Stayton, P. S. (2006) Controlling the aggregation of conjugates of streptavidin with smart block copolymers prepared via the RAFT copolymerization technique. Biomacromolecules 7, 2736–2741. 6. Sengupta, S., Raja, K. S., Kaltgrad, E., Strable, E., and Finn, M. G. (2005) Virus-glycopolymer conjugates by copper(I) catalysis of atom transfer radical polymerization and azide-alkyne cycloaddition. Chem. Commun. 34, 4315–4317. 7. Sengupta, S., Kuzelka, J., Singh, P., Lewis, W. G., Manchester, M., and Finn, M. G. (2005) Accelerated bioorthogonal conjugation: a practical method for the ligation of diverse functional molecules to a polyvalent virus scaffold. Bioconjugate Chem. 16, 1572–1579. 8. Hou, S., Sun, X. L., Dong, C. M., and Chaikof, E. L. (2004) Facile synthesis of chain-end functionalized glycopolymers for site-specific bioconjugation. Bioconjugate Chem. 15, 954–959.

9. Vazquez-Dorbatt, V., and Maynard, H. D. (2006) Biotinylated glycopolymers synthesized by atom transfer radical polymerization. Biomacromolecules 7, 2297–2302. 10. Lele, B. S., Murata, H., Matyjaszewski, K., and Russell, A. J. (2005) Synthesis of uniform protein-polymer conjugates. Biomacromolecules 6, 3380–3387. 11. De, P., Li, M., Gondi, S. R., and Sumerlin, B. S. (2008) Temperature-regulated activity of res­ ponsive polymer-protein conjugates prepared by grafting-from via RAFT polymerization. J. Am. Chem. Soc. 130, 11288–11289. 12. Bontempo, D., Heredia, K. L., Fish, B. A., and Maynard, H. D. (2004) Cysteine-reactive polymers synthesized by atom transfer radical polymerization for conjugation to proteins. J. Am. Chem. Soc. 126, 15372–15373. 13. Bontempo, D., and Maynard, H. D. (2005) Streptavidin as a macroinitiator for polymerization: in situ protein-polymer conjugate formation. J. Am. Chem. Soc. 127, 6508–6509. 14. Shi, W., Dolai, S., Averick, S., Fernando, S. S., Saltos, A. J., L’Amoreaux, W., Banerjee, P. and, and Raja, K. S. (2009) A general methodology toward drug/dye incorporated living copolymer-protein hybrids: (NIRF dye-glucose) copolymer-avidin/BSA conjugates as prototypes. Bioconjugate Chem. 20, 1595–1601. 15. Shi, W., Dolai, S., Rizk, S., Hussain, A., Tariq, H., Averick, S., L’Amoreaux, W., El Idrissi, A., Banerjee, P. and, and Raja, K. S. (2007) Synthesis of monofunctional curcumin derivatives, clicked curcumin dimer, and a PAMAM dendrimer curcumin conjugate for therapeutic applications. Org. Lett. 9, 5461–5464. 16. Kolb, H. C., Finn, M. G., Sharpless, K. B. (2001) Click Chemistry: Diverse Chemical Function from a Few Good Reactions. Angew. Chem. Int. Ed. 40, 2004–2021.

Chapter 4 Dye/DNA Conjugates as Multiple Labels for Antibodies in Sensitive Fluorescence Immunoassays Qin Zhang, Shengchao Zhu, and Liang-Hong Guo Abstract Fluorescence immunoassays are widely used in life science research, medical diagnostics, and environmental monitoring due to the intrinsically high specificity, simplicity, and versatility of immunoassays as well as the availability of a large variety of fluorescent labeling molecules. However, the sensitivity of immunoassays needs to be improved further to meet the ever-increasing demands of the new proteomics era. We have developed a novel and simple method to increase immunoassay sensitivity by attaching multiple fluorescent labels on an antibody with a dye/DNA conjugate. Our strategy is to use a DNA fragment as a molecular carrier to attach multiple fluorescent dyes to an antibody at a single site. The dye/DNA conjugate is not presynthesized, but rather formed in situ as part of the immunoassay. Our results demonstrate that by using a 219-bp DNA fragment in conjunction with SYBR Green I fluorescent DNAbinder, the sensitivity of both direct and competitive fluorescence immunoassays is improved by orders of magnitude, reaching a lower detection limit of 1.9 pg/mL for 17b-estradiol. Key words: Fluorescence immunoassay, Antibody labeling, Multiple labels, Dye/DNA conjugate

1. Introduction Immunoassays are widely used in life science research, medical diagnostics, and environmental monitoring due to their high specificity, simplicity, and versatility. The acquirement of detection signal usually entails labeling of antibodies or antigens with specific types of signal molecules or groups, such as radioisotopes, fluorescent dyes, metal complexes, and enzymes (1–3). For different types of labels, there are different detection methods such as colorimetric (3), chemiluminescent, electrochemiluminescent, and amperometric (4–6) detection. Each label and associated detection method has its own pros and cons. For example, although chemiluminescence is a highly sensitive detection Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_4, © Springer Science+Business Media, LLC 2011

43

44

Zhang, Zhu, and Guo

method, it is not suitable for microarray detection, since the light signal must be produced in the liquid phase and the lightemitting species is not confined to the surface (7). Furthermore, while immuno-PCR provides an amazing improvement in immunoassay sensitivity up to 10,000-fold (8), a series of complex and separate procedures need to be carried out for DNA amplification and quantification. Fluorescence thus remains a desirable method to satisfy the demands of protein microarray detection in terms of label stability, assay robustness, detection sensitivity, and multiplexity. In traditional fluorescence immunoassays, the antibody is labeled with a very limited number of fluorescent dyes such as FITC, Cy-3, and Cy-5 so as to minimize nonspecific binding and loss of antibody reactivity. Consequently, these immunoassays suffer from inferior sensitivity. In recent years, organic dendrimers (9), quantum dots (10), fluorophore-doped silica (11), and latex beads (12) have been employed as multiple labels on an antibody or antigen to improve immunoassay sensitivity. Although sensitivity is enhanced tremendously, these materials bear some problems such as high cost, nonuniformity, poor dispersion in aqueous solutions, and difficulty in protein attachment. We have developed a simple and easy method of attaching multiple fluorescent labels at a single site on antibodies (13). Our strategy is to first link a DNA sequence to an antibody, and then use the DNA as a molecular carrier to attach a large number of DNA-binding fluorescent dyes to the antibody (Fig. 1.). We have demonstrated the utility of the dye/DNA conjugate label in both direct and competitive fluorescence immunoassays. In a direct immunoassay for the detection of goat anti-mouse antibody, mouse IgG was adsorbed to the surface of a 96-well plate, and was recognized by a biotinylated goat anti-mouse antibody in

Fig. 1. (a) Schematic illustration of the antibody multiple labeling strategy using a biotinterminated DNA to carry a large number of fluorescent DNA binders (SYBR Green I), and streptavidin to link the DNA to a biotinylated antibody. (b) Schematic illustration of the conventional approach of using FITC-streptavidin to label an antibody. Reproduced with permission from ref. 14 © 2008 from Elsevier B.V.

Dye/DNA Conjugates as Multiple Labels for Antibodies

45

solution. A 219-bp double-stranded oligonucleotide terminated with a biotin was linked to the antibody through the biotin/ streptavidin/biotin interaction. Multiple labeling of the antibody was achieved after SYBR Green I (a fluorescent DNA binder) was added into the solution and bound to the oligonucleotide at high ratios. By comparison with fluorescein-labeled streptavidin, the assay with the dye/DNA label achieved a 150-fold lower detection limit (14). In a competitive immunoassay for the detection of 17b-estradiol, the antigen was again firstly adsorbed to a 96-well plate. Biotinylated anti17b-estradiol antibody and free 17b-estradiol were added into the well. After the immunoreaction, the antibody was multiply labeled with the dye/DNA conjugate by the procedure described above. In this case, the detection limit was lowered by approximately 200-fold over that achievable with fluoresceinlabeled streptavidin, and reached 1.9 pg/mL 17b-estradiol (14). The novel multiple-labeling method described herein uses readily available chemical and biochemical reagents, and is simple to implement. It thus holds great potential for the sensitive detection of protein microarrays.

2. Materials 2.1. Preparation of Biotinylated and FITC-Labeled Streptavidin and Antibody

1. BT-NHS solution: 20  mg Biotinyl-N-hydroxysuccinimide (BT-NHS) (Sigma, St. Louis, MO) is dissolved in 10 mL dry DMF (N,N-dimethylformamide). The solution is freshly prepared for each labeling experiment (see Note 1). 2. FITC solution: 50  mg Fluorescein isothiocyanate (FITC) (Amresco, Solon, OH) is dissolved in 10 mL dry DMSO (analytical purity) in a dark tube. The solution is freshly prepared for each labeling experiment (see Note 2). 3. Carbonate buffer: 50  mM Na2CO3, 50  mM NaHCO3, pH 9.15 (adjusted with HCl). 4. Protein storage buffer: 137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, 1.5 mM KH2PO4 (adjust to pH 7.4 with HCl if necessary), containing 0.1% (w/v) BSA and 0.1% (w/v) NaN3. Store at 4°C. 5. D-Salt™ polyacrylamide desalting column (Pierce, Rockford, IL). Store at 4°C. 6. Desalting column elution buffer: 137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, 1.5 mM KH2PO4, pH 7.4 (see Note 3). 7. Desalting column storage buffer: 0.05 M NaCl, 0.02% (w/v) NaN3. 8. Microcon YM-3 centrifugal filter device (molecular weight cutoff: 3,000 kDa) (Millipore, Bedford, MA). 9. HABA/avidin reagent (Sigma, St. Louis, MO). Store at 4°C.

46

Zhang, Zhu, and Guo

2.2. Preparation of Biotinylated Double-Stranded DNA Oligonucleotide

1. Design of short-strand DNA oligonucleotides: A 30-bp BT-­DNA (biotinylated double-stranded DNA oligonucleotide) fragment is designed according to the principles of primer design (see Note 4). Strand A: biotin-5¢-TTT TTT TTT GCG GGT AAC GTC AAT ATT AAC TTT ACT CCC-3¢; strand B: 5¢GGG AGT AAA GTT AAT ATT GAC GTT ACC CGC-3¢. 2. PCR primer design for long-strand DNA oligonucleotides: a pair of PCR primers is designed according to the principles of primer design in order to obtain a 219-bp BT-DNA (see Note 5). 3. Hybridization buffer (2× SSC): 0.3 M NaCl, 30 mM sodium citrate, pH 7.0. 4. 1× PCR mixture: 5 mL 10× PCR buffer (Takara Bio, Japan), 1.25  U Taq polymerase, 0.2  mM dNTP mixture, 0.4  mM each of primers, 10  ng template DNA, 1.5  mM MgCl2. Prepare a 1× PCR mixture freshly for each experiment. 5. DNA gel electrophoresis buffer (0.5× TBE): 45  mM Tris– borate, 1 mM EDTA, pH 8.0. 6. DNA fragment purification kit (Takara Bio). 7. DNA storage buffer (10× TE): 10  mM Tris–HCl, 1  mM EDTA, pH 8.0.

2.3. Optimization of Fluorescent Immunoassay Using the Dye/DNA Conjugate Labels

1. Coating buffer: 50 mM Na2CO3, 50 mM NaHCO3, pH 9.6 (adjusted with HCl). 2. Tris–NaCl buffer: 50 mM Tris, 50 mM NaCl, adjust to pH 8.0 with HCl. 3. Blocking buffer: 1% (w/v) BSA in Tris–NaCl buffer. Store at 4°C (see Note 6). 4. Reaction buffer: 0.1% (w/v) BSA in Tris–NaCl buffer. Store at 4°C. 5. Washing buffer: 0.1% (v/v) Tween 20 is added into Tris– NaCl buffer and stored at 4°C. A fresh solution is prepared for each experiment. 6. Detection buffer: 50  mM Tris–HCl and 50  mM NaCl, adjusted to pH 8.0 with HCl. 7. SYBR Green I solution: SYBR Green I (SG1) is dissolved in DMSO and its concentration determined using a UV–visible spectrometer. The dye solution is stable for up to half a year if stored at −20°C in the dark. SG1 dissolved in DMSO is diluted freshly with the detection buffer to the desired concentration (see Note 7). 8. 17b-Estradiol 6-(O-carboxymethyl) oxime:BSA (E2-BSA) (Sigma, St. Louis, MO). Dissolve in protein storage buffer at the desired concentration and store at −20°C. A working solution of E2-BSA is diluted freshly with the coating buffer. 9. Luminescence spectrometer.

Dye/DNA Conjugates as Multiple Labels for Antibodies

47

3. Methods 3.1. Preparation of Biotinylated Goat Anti-mouse Antibody

1. React goat anti-mouse antibody with BT-NHS at a molar ratio of 5.6 BT-NHS:1 antibody in carbonate buffer for 4 h at room temperature (RT) with magnetic stirring. 2. After the reaction, unbound BT-NHS and the antibody are separated by gel filtration on a D-Salt™ polyacrylamide desalting column (see Note 8). 3. Antibody concentration is determined by measuring the absorbance at 280 nm on a UV–visible spectrophotometer. 4. Optional: If necessary, the collected antibody solution is concentrated in a microconcentrator (Microcon YM-3) by centrifuging at 12,000 × g for 30 min at 4°C. The final antibody concentration is determined as above. 5. The biotin labeling ratio on the antibody is determined by using the HABA/avidin reagent (see Note 9). 6. The prepared biotinylated goat anti-mouse antibody (BT-Ab) sample should be aliquoted and stored at −20°C.

3.2. Preparation of FITC-Labeled Streptavidin and Antibody

1. Mix the antibody or streptavidin with FITC at a molar ratio of 5.6 FITC:1 protein in carbonate buffer for 4 h at RT with magnetic stirring (see Note 10). 2. After the reaction, the remaining free FITC is separated from the protein by gel filtration on a D-Salt™ polyacrylamide desalting column. 3. Protein and FITC concentrations are determined by measuring the absorbance at 280 and 494  nm, respectively, on a UV–visible spectrophotometer (see Note 11). 4. Optional: If necessary, the collected protein solution is concentrated in a microconcentrator (Microcon YM-3) by centrifuging at 12,000 × g for 30  min at 4°C. The final protein and FITC concentrations are determined as above. 5. FITC-labeled protein samples should be aliquoted and stored in the dark at −20°C.

3.3. Preparation of Biotinylated Short Double-Stranded DNA Oligonucleotides

1. The two 30-bp complementary oligonucleotides are dissolved in 2× SSC buffer, heated at 95°C for 5 min on a thermocycler, and then allowed to naturally cool down to room temperature. 2. The concentration of the hybridized oligonucleotides is determined by measuring the absorbance at 260 nm. 3. The prepared BT-DNA should be aliquoted and stored at −20°C.

48

Zhang, Zhu, and Guo

3.4. Preparation of Biotinylated Long Double-Stranded DNA Oligonucleotides

1. A tube containing a freshly prepared PCR mixture is loaded into a thermocycler. 2. PCR: Two 219-bp complementary oligonucleotides are heated at 94°C for 3 min, followed by 30 PCR cycles. Each PCR cycle consists of a denaturation at 94°C for 30 s, annealing at 56°C for 30 s, and extension at 72°C for 1 min. After 30 PCR cycles, the mixture is reacted at 72°C for 3 min for the final chain extension. The solution is then ramped down to 4°C and held at that temperature until sample analysis (see Note 12). 3. To verify the desired product, the PCR mixture is separated by electrophoresis on a 1.2% (w/v) agarose gel in 0.5× TBE buffer. The gel is stained with 0.5 mg/mL ethidium bromide and visualized under UV illumination. An example of the results produced is shown in Fig. 2. 4. The PCR product is purified with a DNA fragment purification kit.

Fig. 2. Gel electrophoresis image of a 219-bp long double-stranded DNA oligonucleotide PCR product. Lanes 1 and 2 : PCR product; lane 3 : DNA molecular weight standards (DL2000 DNA markers). The PCR product was loaded onto a 1.2% agarose gel in TBE buffer, and the gel was stained with 0.5  mg/mL ethidium bromide. Reproduced with permission from ref. 14 © 2008 from Elsevier B.V.

Dye/DNA Conjugates as Multiple Labels for Antibodies

49

3.5. Optimization of Fluorescent Immunoassays Using the Dye/DNA Conjugate Labels

1. A white 96-well plate is coated with 100 mL of FITC-labeled mouse IgG dissolved at various concentrations in coating buffer by incubating overnight at 4°C.

3.5.1. Optimization of Mouse IgG Concentration

3. The plate is blocked with 300  mL of blocking buffer overnight at 4°C, and then washed as above.

2. The plate is then washed three times with 300 mL of washing buffer.

4. Fluorescence intensity is measured on a luminescence spectrometer with 495 nm excitation, 525 nm emission, 5 nm slit width, and 515 nm cutoff filter. Based on the fluorescence intensity vs. IgG concentration curve, the optimal mouse IgG concentration can be obtained. In this example protocol, the optimal mouse IgG concentration was determined to be 30 mg/mL.

3.5.2. Optimization of BT-Ab Concentration

1. A white 96-well plate is coated with 100  mL of 30  mg/mL mouse IgG solution. 2. 100  mL of FITC-Ab of different solution concentrations is added into the well and reacted at 37°C for 2 h with shaking. 3. After washing, fluorescence is measured as described above. In this example protocol, the optimal antibody concentration of 20  mg/mL is obtained from the fluorescence intensity vs. FITC-Ab concentration curve.

3.5.3. Optimization of Streptavidin Concentration

1. After plate coating with the optimal concentration of mouse IgG (Subheading 3.5.1) and immunoreaction with the optimal concentration of BT-Ab (Subheading 3.5.2), 100 mL of FITC-SA of different concentrations is added into the well and reacted at 37°C for 2 h. 2. Fluorescence intensity is then measured, and the optimal concentration of streptavidin is obtained as above. In this example protocol, the optimal concentration of streptavidin of 5 mg/mL is obtained.

3.5.4. Optimization of BT-DNA and Dye Concentration

1. After plate coating with the optimal concentration of mouse IgG, immunoreaction with the optimal concentration of BT-Ab, and affinity reaction with the optimal concentration of streptavidin, 100 mL BT-DNA of different concentrations is added to the well and reacted at 37°C for 1 h with shaking. 2. After washing, 100  mL SYBR Green I (SG1) of different concentrations is added into the plate and incubated for 10 min at RT with shaking. 3. The fluorescence intensity is measured and the optimal concentration of BT-DNA and SG1 is determined.

50

Zhang, Zhu, and Guo

In this example protocol, the optimal concentration of BT-DNA was determined to be 88 nM, and the optimal concentration of SG1 was determined to be 574 nM. 3.6. Direct Immunoassay of Goat Anti-mouse Antibody Using the Dye/DNA Conjugate Labels

1. A white 96-well plate is coated with 100  mL of 30  mg/mL mouse IgG by incubating overnight at 4°C, and washed three times with washing buffer. 2. The plate is then blocked with 300 mL of 1% BSA in PBS, pH 7.4 by incubating overnight at 4°C, and washed as above. 3. 100  mL BT-Ab of various concentrations is added into the IgG coated well and incubated at 37°C for 2 h with shaking. The plate is again washed. 4. 100 mL of 5 mg/mL streptavidin is added into the well and reacted at 37°C for 2  h with shaking, followed by plate washing. 5. BT-DNA (88 nM) is added and incubated at 37°C for 1  h with shaking, followed by plate washing. 6. SYBR Green I is added into the plate and reacted for 10 min at RT with shaking, followed by plate washing. 7. Fluorescence intensity is measured with 495  nm excitation, 520 nm emission, a 5-nm slit width, and a 515-nm cutoff filter. 8. The lower limit of detection (LOD) is calculated from the concentration curve. An example concentration curve is shown in Fig. 3.

Fig. 3. Fluorescence immunoassay for the detection of goat anti-mouse IgG on mouse IgG coated 96-well plates using the 219-bp DNA/SYBR Green I conjugate label (filled diamond ), or the 30 bp oligonucleotide/SYBR Green I conjugate label (filled triangle). Each data point is the average of three replicate measurements. Reproduced with permission from ref. 14 © 2008 from Elsevier B.V.

Dye/DNA Conjugates as Multiple Labels for Antibodies

3.7. Competitive Immunoassay of 17b-Estradiol Using the Dye/DNA Conjugate Labels

51

1. A white 96-well microplate is coated with 100 mL of 10 mg/ mL E2-BSA by incubating overnight at 4°C, and washed (see Note 13). 2. The plate is then blocked with 300 mL of the blocking buffer by incubating overnight at 4°C, and washed. 3. 50 mL of 20 mg/mL biotinylated E2 antibody and 50 mL E2 of various concentrations are added into the coated well, incubated at 37°C for 2 h with shaking, and then the plate is washed (see Note 14). 4. 100 mL of 5 mg/mL streptavidin is added into the well and incubated at 37°C for 2  h with shaking. The plate is then washed. 5. 88 nM biotin-labeled 219 bp DNA is added into the well and reacted at 37°C for 1  h with shaking. The plate is then washed. 6. 574 nM SG1 is added into the well and reacted for 7 min at RT with shaking, followed by plate washing. 7. Fluorescence intensity is measured under the same conditions as described above in Subheading 3.6 and the LOD is calculated from the competition curve (see Note 15). An example of the competition curve is shown in Fig. 4.

4. Notes 1. Because the NHS group is very easily hydrolyzed, BT-NHS should be stored frozen in an air-tight container with desiccants. For the same reason, silica gel desiccants are used to keep the DMF solvent dry. 2. To avoid photolysis, FITC should be stored in the dark. 3. The composition of the desalting column elution buffer depends on the storage buffer for the labeled protein. Typically, adding 0.05  M NaCl is helpful in the elution process. 4. The design of short-sequence DNA oligonucleotides conforms to several principles. (1) There is no special restriction to the type of bases; however, intrachain hairpins must be avoided. (2) The length is one of the decisive factors in assay sensitivity. The appropriate range of lengths is between 20 and 60 bp, which can bind enough DNA probes and also reduce spatial hindrance and the cost of nucleic acid synthesis. (3) In this example protocol, Oligo d(T)9 is a homo-oligomeric deoxyribonucleotide used as a linker to reduce spatial hindrance between biotin and streptavidin binding.

52

Zhang, Zhu, and Guo

Fig. 4. (a) Competitive fluorescence immunoassay for the detection of E2 on E2-BSAcoated 96-well plates using the long chain DNA/dye conjugate-SA labeled anti-E2 antibody (filled diamond  ) and FITC-SA labeled anti-E2 antibody (filled square). (b) An expanded view of (a) at low E2 concentrations. Each data point is the average of three replicate measurements. Reproduced with permission from ref. 14 © 2008 from Elsevier B.V.

5. The principles of primer design for long DNA oligonucleotides are the same as that for short DNA oligonucleotides. In this example protocol, a 219-bp DNA sequence is synthesized by PCR and used in fluorescence immunoassays. Whether even longer DNA sequences could improve fluorescence sensitivity is unknown. If the sequence is too long, a random coil may form and disturb the binding of the DNA with fluorescent dyes.

Dye/DNA Conjugates as Multiple Labels for Antibodies

53

6. Other blocking reagents such as 1% (w/v) gelatin can also be used. 7. SYBR Green I was chosen in our work due to its high DNAbinding affinity and excellent fluorescence characteristics. If necessary, other DNA-binding fluorescent dyes such as SYTOX Orange may also be used. However, the optimal concentration and reaction conditions will need to be optimized accordingly. 8. Because both BT-NHS and antibody are colorless, the volume of each collected fraction should be kept as low as possible for good resolution. The collected fractions that contain protein are identified by measuring the absorbance of each fraction at l = 280 nm. 9. According to the manufacturer’s instructions (available from Sigma), the biotin content can be calculated based on the binding of the dye HABA to avidin and the ability of biotin to displace the dye in stoichiometric proportions. 10. The general procedure is the same as that for BT-NHS labeling (Subheading 3.1). Note, however, that all the reactions are conducted in the dark. 11. Because FITC also absorbs at 280 nm, the measured absorbance at 280 nm needs to have the FITC contribution subtracted out before the protein concentration can be calculated correctly. 12. The stated PCR conditions were optimized from a series of pilot studies. If the primer and target sequence are changed, the PCR conditions may need to be adjusted. 13. The optimal concentration of E2-BSA to be used is determined by performing experiments similar to those conducted to determine the optimal mouse IgG coating concentration (Subheading 3.5.1). 14. The concentration of biotinylated E2 antibody employed is the EC50 value (the antibody concentration producing half of the peak value of fluorescence intensity) in the fluorescence vs. antibody concentration curve. 15. All the results shown are from at least three replicates. The lower LOD is calculated based on three times the standard deviation above the blank value.

Acknowledgments This work was supported by the Knowledge Innovation Program of the Chinese Academy of Sciences (KSCX2-YW-G-059) and the National Basic Research Program of China (2006CB403303).

54

Zhang, Zhu, and Guo

References 1. Berson, S. A., Yalow, R. S. (1959) Assay of plasma insulin in human subjects by immunological methods. Nature 184, 1648–1649. 2. Li, T. M., Parrish, R. F. (2002) Fluorescence and Immunodiagnostic Methods. In Topics in Fluorescence Spectroscopy. Volume 3: Biochemical Applications. Lakowicz, J. R. (Ed.). Springer US, New York. pp 273–287. 3. Blackburn, G. F., Shah, H. P., Kenten, J. H., Leland, J., Kamin, R. A., Link, J., Peterman, J., Powell, M. J., Shah, A., Talley, D. B. (1991) Electrochemiluminescence detection for development of immunoassays and DNA probe assays for clinical diagnostics. Clin. Chem. 37, 1534–1539. 4. Marquette, C. A., Blum, L. J. (2006) State of the art and recent advances in immunoanalytical systems. Biosens. Bioelectron. 21, 1424–1433. 5. Kricka, L. J. (1999) Chemiluminescence and Bioluminescence. Anal. Chem. 71, 305-308. 6. Bange, A., Halsall, H. B., Heineman, W. R. (2005) Microfluidic immunosensor systems. Biosens. Bioelectron. 20, 2488–2503. 7. Schweitzer, B., Predki, P., Snyder, M. (2003) Microarrays to characterize protein interactions on a whole-proteome scale. Proteomics 3, 2190–2199. 8. Ong, K. K., Jenkins, A. L., Cheng, R., Tomalia, D. A., Durst, H. D., Jensen, J. L., Emanuel, P. A., Swim, C. R., Yin, R. (2001)

Dendrimer enhanced immunosensors for biological detection. Anal. Chim. Acta 444, 143–148. 9. Zhou, M., Roovers, J., Robertson, G. P., Grover, C. P. (2003) Multilabeling Biomo­ lecules at a Single Site. 1. Synthesis and Characterization of a Dendritic Label for Electrochemiluminescence Assays. Anal. Chem. 75, 6708–6717. 10. Chan, W. C. W., Nie, S. (1998) Quantum dot bioconjugates for ultrasensitive nonisotopic detection. Science 281, 2016–2018. 11. Wang, L., Wang, K., Santra, S., Zhao, X., Hilliard, L. R., Smith, J. E., Wu, Y., Tan, W. (2006) Watching silica nanoparticles glow in the biological world. Anal. Chem. 78, 646–654. 12. Bangs, L. B. (1996) New developments in particle-based immunoassays: Introduction. Pure Appl. Chem. 68, 1873–1879. 13. Zhang, Q., Guo, L. H. (2007) Multiple labeling of antibodies with Dye/DNA conjugate for sensitivity improvement in fluore­­ scence immunoassay. Bioconjugate Chem. 18, 1668–1672. 14. Zhu, S. C., Zhang, Q., Guo, L. H. (2008) Part-per-trillion level detection of estradiol by competitive fluorescence immunoassay using DNA/dye conjugate as antibody multiple labels. Anal. Chim. Acta 624, 141–146.

Chapter 5 Chemoselective Modification of Viral Proteins Bearing Metabolically Introduced “Clickable” Amino Acids and Sugars Partha S. Banerjee and Isaac S. Carrico Abstract The inherent difficulty of performing chemical modifications of proteins in a truly site-specific fashion is often compounded by the need to work within complex biological settings. In order to alleviate this complication, targets can be “prelabeled” metabolically with unnatural residues, which allow access to highly selective bioorthogonal reactions. Due to their small size, permissibility within biosynthetic pathways and access to reactions with high specificity, azides provide excellent bioorthogonal handles. This two-step labeling process is emerging as a highly effective means to modify therapeutic proteins. In this chapter, we take this strategy a step further and apply chemoselective ligation to remodel the surfaces of adenoviruses. Despite the large number of ongoing clinical trials involving these complex mammalian viruses, new methods for their facile, flexible surface modification are necessary to drive the development of next-generation therapeutics. Here we demonstrate the modification of azides on adenoviral surfaces via a straightforward chemoselective protocol based on copper-assisted “click” chemistry. This method provides access to a wide array of effector functionalities without sacrificing infectivity. Key words: Chemoselective modification, Bioorthogonal, CuAAC, “Click” chemistry, Azide, Unnatural amino acid, Azido sugar, Alkyne probe, Adenovirus

1. Introduction The specific chemical modification of proteins can enable fundamental physiological and biochemical studies as well as allow the production of next-generation therapeutics. While a number of strategies allow selective chemical modification, one of the most conceptually straightforward approaches is the introduction of selectively reactive functional groups such as azides, aldehydes, alkynes, etc., within a target protein (1–3). This can be accomplished using a number of methodologies, including synthetic/semisynthetic Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_5, © Springer Science+Business Media, LLC 2011

55

56

Banerjee and Carrico

protein synthesis, enzymatic modification, and the metabolic introduction of unnatural amino acid resides. The latter method has the distinct advantage of enabling the production of modifiable proteins within the native cellular environment (4). In addition, the metabolic introduction of unnatural amino acids and monosaccharides can demonstrate high efficiency and does not require the use of engineered cell lines (5). The resulting flexibility thus allows the exploration of new target proteins of interest quickly regardless of cell type. However, the incorporation of unnatural residues into cellular, nontarget proteins has the potential to perturb cellular physiology. Mammalian viruses have tremendous therapeutic potential as gene delivery, oncolytic and vaccine agents (6); but unfortunately, the inability to efficiently tune their targeting properties and modulate the host immune response has limited their success in these areas. Genetic methods have been widely used to modify viral properties, but in many instances such types of manipulations have led to a propensity for loss in infectivity and limited access to effector functionality (7, 8). In order to sidestep these issues, we have developed an approach to specifically introduce unnatural amino acids and monosaccharides onto the surface of mature, infectious, adenovirus particles. These modified viral vectors demonstrate no loss in either production or infectivity, and can be selectively modified with a wide range of functionalities.

2. Materials All chemicals and reagents were obtained from commercial sources and used without further purification unless otherwise noted. 2.1. Cell Culture, Virus Infection, and Harvesting

1. 100-mm cell culture dishes, sterile. 2. DMEM complete medium: Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% (v/v) bovine calf serum. 3. DMEM (−Met/−Cys): DMEM minus l-methionine, minus l-cysteine, supplemented with 10% (v/v) bovine calf serum. 4. AHA stock solution (25×): l-Azidohomoalanine (AHA) (Invitrogen, Carlsbad, CA) (see Note 1) is dissolved in DMEM (−Met) medium to make a 100-mM solution. Ideally, the solution should be made fresh before use, but it can also be stored at 4°C overnight. 5. Methionine stock solution (25×): l-methionine is dissolved in DMEM (−Met) medium to make a 100-mM solution. Make fresh before use.

Chemoselective Modification of Viral Proteins

57

6. Cysteine stock solution (100×): l-cysteine dissolved in DMEM (−Met) medium to make a 200-mM solution. Make fresh before use. 7. AHA labeling medium: DMEM (−Met/−Cys) supplemented with AHA stock solution and cysteine stock solution to a final concentration of 4 mM AHA and 2 mM cysteine. 8. Methionine (control) labeling medium: DMEM (−Met/−Cys) supplemented with methionine stock solution and cysteine stock solution to a final concentration of 4 mM methionine and 2 mM cysteine. 9. Ac4GalNAz stock solution (1,000×): Peracetylated N-azidoacetylgalactosamine (Ac4GalNAz) (Invitrogen) (see Note 1) is dissolved in methanol at a concentration of 50 mM and stored at 4°C. 10. Adenovirus type 5 (Ad5) containing a GFP or luciferase transgene (Vector BioLabs, Eagleville, PA). Maintain the viruses at −20°C in 5 mM Tris–HCl buffer (pH 8.0) containing 0.5  mM MgCl2, 50  mM NaCl, 25% (v/v) glycerol, and 0.05% (w/v) BSA. 11. Human embryonic kidney (HEK) 293 cells, maintained in DMEM supplemented with 10% (v/v) bovine calf serum, 2  mM glutamine, 100  U/mL penicillin, and 100  mg/mL streptomycin (see Notes 2 and 3). 12. TD buffer: 137 mM NaCl, 5 mM KCl, 0.7 mM Na2HPO4, and 25 mM Tris–HCl, pH 7.5. Store at room temperature (RT). 13. TC solution (200×): 180  mM CaCl2 and 210  mM MgCl2. Store at RT. 14. Cesium chloride (CsCl) solutions of 1.25 g/mL, 1.35 g/mL, and 1.40 g/mL in TD buffer. Sterile filter and store at RT. 15. Beckman Ultra Clear 3.5-in. centrifuge tubes. 16. Beckman ultracentrifuge equipped with a SW 41 and SW 60 rotor. 17. Virus storage buffer: PBS (pH 7.2) containing 0.5 mM CaCl2, 0.9 mM MgCl2, and 10% (v/v) glycerol. Store at −20°C. 2.2. Azide–Alkyne Cycloaddition or “Click” Chemistry and Virus Purification

1. Copper(I) bromide (CuBr) is dissolved in deoxygenated dimethylsulfoxide (DMSO) to make a 50-mM stock solution. Make fresh every time before use. 2. 2 M Tris–HCl buffer, pH 8.0. Store at RT. 3. Bathophenanthroline disulfonic acid disodium salt (MP Biomedicals, Solon, OH) is dissolved in water to make a 50-mM stock solution. Store at RT. 4. 10  mM Alkyne-TAMRA probe [also known as Click-iT® Tetramethylrhodamine (Invitrogen)] in water. Store at −20°C.

58

Banerjee and Carrico

5. Virus storage buffer: PBS (pH 7.2) containing 0.5 mM CaCl2, 0.9 mM MgCl2, and 10% (v/v) glycerol. Store at −20°C. 6. Centri-Sep™ spin columns (Princeton separations, Adelphia, NJ). 7. Ethylenediaminetetraacetic acid (EDTA). 2.3. SDSPolyacrylamide Gel Electrophoresis

1. Laemmli sample buffer (2×): 125  mM Tris–HCl (pH 6.8), 4% (w/v) SDS, 20% (v/v) glycerol, 10% (v/v) 2-mercaptoethanol, and 0.004% (w/v) bromophenol blue. 2. Resolving buffer (4×): 1.5 M Tris–HCl, pH 8.7. Store at RT. 3. Stacking buffer (8×): 1 M Tris–HCl, pH 6.8. Store at RT. 4. 40% (w/v) acrylamide/bisacrylamide solution (29:1, w/w) (3.3% Cross-linker). 5. N,N,N,N ¢-tetramethylethylenediamine (TEMED). 6. 10% (w/v) SDS solution in water. Store at RT. 7. 10% (w/v) Ammonium persulfate solution in water. Prepare fresh. 8. Running buffer (1×): 187  mM glycine, 19  mM Tris–HCl, 3.5 mM SDS in water. Store at 4°C. 9. Prestained protein molecular weight markers. 10. Fluorescent gel scanner equipped with the required excitation and emission filters. For imaging TAMRA-labeled adenovirus particles, we used the following filter setup: Ex 532 nm; Em 580 nm, BP 30.

3. Methods In order to diminish the level of toxicity due to the incorporation of unnatural amino acids within cellular proteins, metabolic labeling of adenovirus with AHA (Subheading 3.1) is carried out over a 6-h time period during the viral growth phase. This was done expressly to maximize overlap with adenoviral structural protein expression, specifically 18–24  h postinfection (9). Although we have not explored this, it is expected that longer labeling times may lead to reduced levels of virus production and infectivity. Although metabolic labeling with peracetylated sugars has demonstrated cellular toxicity (10, 11), the concentrations needed for the efficient labeling of adenovirus particles (Subheading 3.2) are well below toxic levels. Furthermore, the metabolic conversion of Ac4GalNAz to the donor sugar, UDP-GlcNAz, is relatively slow. As a result, we chose to incubate the infected host cells with Ac4GalNAz throughout the viral production period. It is worth noting that no losses in either the viral production yield or infectivity

Chemoselective Modification of Viral Proteins

59

of the resulting particles was observed using either of the metabolic labeling protocols described in this chapter. The kinetics of chemical modification via copper-catalyzed azide–alkyne cycloaddition (CuAAC) are not well understood and can vary greatly with changes to the composition of the catalyst and the concentrations of either reactant. Notably, under the conditions described herein (Subheading  3.4), this “click” reaction proceeds with high yield (~90%) within a 12-h period. However, incomplete removal of the cytotoxic copper catalyst after the ligation reaction can in some instances pose a potential problem, especially for functional assays within animal model systems. Furthermore, it should be noted that the presence of copper may also impede the mass spectral analysis of modified proteins. Two alternative chemistries are available that allow copper-free and highly selective azide modifications. The first, Staudinger ligation, is well characterized and reagents can be obtained commercially; however, this reaction is slower under the described conditions. Alternatively, ring strain-promoted cycloaddition reactions are significantly faster than the Staudinger ligation. Although the cyclooctyne reagents required for strain promoted reactions have been difficult to synthesize, they have recently become commercially available (Jena Bioscience) making this reaction significantly more approachable. Notably, it has been reported that “click” and Staudinger reactions are hampered by the presence of urea and ionic detergents such as SDS, respectively (12). 3.1. Production of Azidoho­moalanineLabeled Adenovirus

1. For virus production, near-confluent 293 cells grown on 100mm culture dishes are used (see Note 2). A batch of ten culture dishes can generally be used for the production of a single type of virus. The cells in each dish of a batch of ten dishes should be treated similarly, and their combined volumes after cell lysis used for purification of the virus. Before infection, determine the cell count by removing the growth medium from the culture dish and loosening the cells with 1  mL of trypsin–EDTA per dish (see Note 3). Count the number of cells using a hemocytometer and calculate the amount of virus required for infection, depending on the average cell count per dish. For infection, use virus at a multiplicity of infection (MOI) of 10 PFU/cell. For normal adenovirus type 5 (Ad5) obtained from commercial or academic sources, the infectivity index – defined as the ratio of infectious virions (determined by the plaque assay method and expressed as PFU per mL) to the total number of physical virus particles per mL – is typically 1:20 but can vary depending on the serotype or source. 2. Prepare a sufficient volume of infection buffer containing 2% (v/v) serum, 1× TC solution, infective Ad5 and TD buffer to give 1 mL per dish (e.g., for 20 dishes, make up to 22 mL of

60

Banerjee and Carrico

infection buffer using TD buffer to bring the solution up to the final total required volume). 3. Near-confluent 293 cells seeded on 100-mm culture dishes are first infected with adenovirus using the prepared infection buffer. Infection is carried out by slow removal of the DMEM complete growth medium, followed by addition of 1 mL of infection buffer per 100-mm dish and incubating at 37° for 1 h with periodic shaking (see Note 4). 4. Cover the cells with 10 mL of fresh DMEM complete medium and incubate at 37°C for 18 h. 5. After 18-h postinfection (see Note 5), carefully remove the growth medium covering the infected cells and wash each culture dish with 5 mL of TD buffer (see Note 6). If necessary for cell-line stability, preincubate the cells with the wash buffer at 37°C for 10 min. 6. Remove the TD buffer wash solution and fill the culture dish with AHA- or methionine (control)-supplemented labeling medium and incubate for 6 h (see Note 5). For 100-mm culture dishes, a minimum of 5 mL of labeling medium per dish should be used (see Note 7). 7. After 24-h postinfection (i.e., 6 h after addition of the labeling medium), remove the supplemental medium carefully (see Note 8) and cover the cells with fresh DMEM complete growth medium (10 mL per dish). Incubate the cells for an additional 18–20 h at 37°C. 3.2. Production of GlcNAz-Labeled Adenovirus

1. Follow steps 1–3 from Subheading 3.1.

3.3. Purification of Metabolically Labeled Adenoviral Particles

1. Remove the cells (which should have become loose and mostly dislodged from the culture dish by this time) from the culture dish, and centrifuge at 2,000 × g and remove excess medium (see Note 9). Resuspend the cell pellet in 8 mL of TD buffer for every ten dishes of infected cells.

2. Cover the cells with 10 mL of fresh DMEM complete growth medium and then add 10 mL of Ac4GalNAz stock solution. Incubate the cells at 37°C for 44 h.

2. Lyse the cells by repeated freeze–thaw cycles (at least three times) in liquid N2. Centrifuge at 2,000 × g to remove cell debris. 3. Prepare a CsCl density gradient for virus purification with 2.0  mL of 1.4  g/mL CsCl at the bottom and 2.5  mL of 1.25 g/mL CsCl on top in an ultracentrifuge tube (see Note 10). Top these with the soluble cell extract obtained from step 2 above and then centrifuge the samples in a swinging bucket rotor at 126,444 × g (average) for 1 h at 15°C in an ultracentrifuge (see Note 11).

Chemoselective Modification of Viral Proteins

61

4. After completion of ultracentrifugation, the virus particles should be visible as a thick white band at the junction of the two CsCl solutions. Make a small perforation at the bottom of the tube with a sterile needle to allow the solution to come out drop by drop, and collect the thick white virus band (see Notes 12–14). 5. Purified virus extracts should be obtained by rebanding of the collected virus sample using 1.35 g/mL of CsCl in an ultracentrifuge spinning at 125,812 × g (average) for 18 h at 15°C (see Notes 15 and 16). The virus can be collected as described above (see Note 17). 6. For storage of the virus particles over long periods of time, dialysis of the collected virus extract should be performed against virus storage buffer. Store the dialyzed virus samples at −20°C with 10% (v/v) glycerol. 3.4. Chemical Labeling of Azide-Modified Virus with Alkyne Probes Using CopperAssisted “Click” Chemistry

While this protocol describes the use of a Cu(I) bathophenanthroline disulfonic acid catalyst under deoxygenating conditions (Fig. 1), the use of other ligands under atmospheric conditions is well documented and provides robust labeling, albeit at slightly lower levels. 1. To a solution of 2 × 1012 viral particles, add Tris–HCl buffer, pH 8.0 at a final concentration of 100 mM (see Note 18). 2. Add bathophenanthroline disulfonic acid disodium salt at a final concentration of 3 mM to the above solution.

Fig. 1. Schematic illustration of virions metabolically modified with unnatural substrates and their subsequent chemical labeling with alkyne probes using two different chemistries, the copper-assisted “click” reaction (top) and the Staudinger ligation reaction (bottom).

62

Banerjee and Carrico

3. To the above mixture, add the desired alkyne probe (e.g., alkyne-TAMRA, a rhodium-based fluorophore dye) at a final concentration of 200–500 mM. 4. Weigh out 7.2 mg of CuBr in an Eppendorf tube and cover it with aluminum foil. 5. Place the reaction mixture obtained from step 3, the Eppendorf tube containing CuBr (step 4), and 1  mL of DMSO inside a deoxygenated glove bag for 6 h. 6. After the 6-h deoxygenation period, dissolve the CuBr in the DMSO (1 mL) to make a 50-mM stock solution. 7. Add a sufficient volume of the CuBr stock solution (step 6) to the reaction mixture to give a final concentration of 1  mM CuBr (see Notes 19 and 20), and allow the labeling reaction to proceed overnight inside the glove bag. 8. Remove the reaction solution from the glove bag and add EDTA solution to a final concentration of 10 mM (see Note 20) to stop the labeling reaction. 9. Purify the labeled virus samples using Centri-Sep™ gel filtration spin columns (see Notes 21 and 22) equilibrated with PBS buffer, pH 7.2 containing 0.5  mM CaCl2, 0.9  mM MgCl2, and 10% (v/v) glycerol. 10. Store the labeled virus at −20°C. 3.5. SDS-PAGE Analysis of TAMRALabeled Adenovirus

The procedures described below are designed for use with the BioRad Mini-PROTEAN® 3 gel electrophoresis system; however, modification to suit any gel apparatus is straightforward. 1. Prepare a 0.75-mm thick 10% (w/v) resolving gel by adding 4.8 mL of H2O, 2.47 mL of 40% (w/v) acrylamide, 2.5 mL of 4× resolving buffer, 100 mL of 10% (w/v) SDS, 100 mL of 10% (w/v) ammonium persulfate, and 5  mL of TEMED. Pour the gel mixture between the glass plates of the MiniPROTEAN® 3 gel cassette while leaving space for the stacking gel, and then cover it with a layer of ethanol. Allow the resolving gel to polymerize for 20–30 min. 2. Prepare the stacking gel by adding 3.6 mL of H2O, 0.623 mL of 40% (w/v) acrylamide, 0.63  mL of 8× stacking buffer, 50 mL of 10% (w/v) SDS, 50 mL of 10% (w/v) ammonium persulfate and 5 mL of TEMED. Pour this gel mixture over the resolving gel and insert the comb. Allow the stacking gel to polymerize for 30 min. 3. Add Laemmli sample buffer (1×) to 20 mL of purified labeled virus and boil for 10 min (see Note 23). Centrifuge the samples for 30 s to remove any insoluble precipitates. 4. Once the stacking gel has polymerized, remove the comb. Attach the gel cassette to the electrophoresis cell and add

Chemoselective Modification of Viral Proteins

63

Fig. 2. SDS-PAGE analysis of TAMRA-labeled Ad5 virions obtained by “click” conjugation of alkyne-TAMRA probe to azide-modified adenovirus particles. (a) Fluorescent gel image of virus particles labeled with alkyne-TAMRA produced at two different concentrations of AHA (4 and 32 mM). Standard concentrations of TAMRA were used to determine the number of dye molecules coupled to each virus particle. Based on comparisons with the protein molecular weight markers, the adenovirus Hexon, Penton, and Fiber capsid proteins appear to be labeled with AHA. (b) Fluorescent gel image of Ac4GalNAz-modified virus particles treated with alkyne-TAMRA under CuAAC conditions. The gel image indicates that only the adenovirus Fiber capsid protein is labeled with GalNAz.

running buffer (see Note 24). Load the required number of wells with the viral protein samples prepared in step 3 and use one well to load the prestained molecular weight markers. Run the gel for 1 h at 200 V. Keep the electrophoresis cell covered with aluminum foil during the entire period (see Note 25). 5. After the tracking dye has run out of the gel (see Note 26), stop the electrophoresis and quickly transfer the gel to a fluorescent gel scanner and scan the gel for fluorescence emission at the appropriate wavelength (based on the fluorophore dye label used) (Fig. 2).

4. Notes 1. Azidohomoalanine (an unnatural azide-functionalized methionine analog) and peracetylated N-azidoacetylgalactosamine (an unnatural azido sugar) can either be purchased from commercial sources or easily synthesized as previously described (13, 14). 2. Before infection, confirm that the cells are 80–90% confluent. Cultures containing overconfluent cells can be difficult to infect and may not produce viruses as desired.

64

Banerjee and Carrico

3. Before infection, count the number to cells using a hemocyto­ meter to calculate the correct amount of Ad5 required for infection. 4. During the 1-h infection period, gently shake the culture dish side to side every 15 min while incubating at 37°C. 5. AHA labeling can be carried out anytime between 12- and 24-h postinfection. Labeling for more than 6  h, however, may result in reduced virion production. 6. Before addition of the labeling medium, wash the cells with TD buffer. Care must be taken at this step, as infected cells are susceptible to becoming washed away. 7. At least 5 mL of labeling medium per 100-mL culture dish is necessary for sufficient metabolic incorporation. The use of lesser amounts of labeling medium tends to leave uncovered cells in the culture dish. 8. During removal of the labeling medium, great care must be taken as infection has proceeded for 24  h and cells at this point are more susceptible to loss during medium removal. To reduce the loss of infected cells, it may not be absolutely necessary to remove the entire labeling medium; complete DMEM medium contains excess methionine that competes away any residual AHA that may remain. 9. Centrifugation of the cells can be carried out at speeds of up to 2,000 × g. The use of higher centrifugation speeds, however, may cause rupture of the infected cells. 10. To create a CsCl density gradient, first add the 1.25 g/mL CsCl solution into the Ultra Clear centrifuge tube. Next, carefully pipette the 1.4 g/mL CsCl solution with the pipette positioned at the bottom of the tube. Take care to release the CsCl solution at this step very slowly; quick release of the heavier CsCl solution may cause mixing of the two liquids. 11. After step 3 in Subheading  3.3, the two liquids and their junction should be clearly visible. 12. After the first ultracentrifugation run, be careful as not to collect cellular proteins that will band above the virus solution. In addition, empty virion capsids can be seen as a lighter layer slightly above the thick white virus band. 13. If excess cellular proteins are present that may interfere with the collection of the virus, use a pipette to remove some of the denser proteins. 14. While collecting the virus band, try to avoid collection of excess CsCl before the virus band starts dripping out. Excess CsCl collected at this step may reduce the sharpness of the virus banding in the subsequent re-centrifugation step.

Chemoselective Modification of Viral Proteins

65

15. During rebanding on 1.35  g/mL of CsCl (step 5 in Subheading 3.3), add the virus collected in the previous step to a 4-mL ultracentrifuge tube and top it with the CsCl solution. After 18 h of centrifugation, a thick white band of pure virus will be visible in the middle of the tube. This rebanding step is necessary to remove empty capsids and other cellular proteins that may have been collected with the virus particles after the first ultracentrifugation run. 16. If for some reason the concentration of virus particles collected after the first ultracentrifugation run is very low, it may be difficult to observe rebanding of the virions after the second 18-h ultracentrifugation run. 17. Care must be taken at these stages during metabolic labeling to minimize the loss of cells and virions. Careful removal of the labeling medium, adding another freeze thaw cycle, careful collection of the first virion band during density gradient ultracentrifugation can all cumulatively add up to give a good yield of the labeled virus. 18. During the deoxygenation step for “click” labeling, keep the virus solution to less than 100 mL in order to minimize losses due to evaporation. 19. Upon addition of CuBr to the reaction mixture, the solution will turn dark green. 20. At the end of the reaction, addition of EDTA should remove the coloration noted above. 21. Purification with Centri-Sep™ spin columns should be carried according to the manufacturer’s instructions provided with the kit. 22. Alternatively, purification can also be carried out by dialysis against virus storage buffer. 23. If concerned about the presence of residual copper after purification, forgo the boiling of the proteins after the addition of sample buffer as boiling of samples in the presence of copper can result in the formation of thick streaks of protein on the SDS-polyacrylamide gel electrophoresis (SDS-PAGE) gel and will cause difficulty during fluorescent visualization. 24. SDS-PAGE running buffer can be reused and stored at 4°C for up to 2 weeks. 25. Cover the electrophoresis apparatus with aluminum foil to reduce photobleaching of the fluorophore label. 26. Allow the tracking dye to run off the gel during electrophoresis, as running the gel longer will result in better resolution and will also allow free fluorophore to run out of the gel.

66

Banerjee and Carrico

References 1. Kiick, K. L., Saxon, E., Tirrell, D. A., and Bertozzi, C. R. (2002) Incorporation of azides into recombinant proteins for chemoselective modification by the Staudinger ligation, Proceedings of the National Academy of Sciences of the United States of America 99, 19–24. 2. Carrico, I. S., Carlson, B. L., and Bertozzi, C. R. (2007) Introducing genetically encoded aldehydes into proteins, Nat Chem Biol 3, 321–322. 3. Kristi, L. K., Jan, C. M. v. H., and David, A. T. (2000) Expanding the Scope of Protein Biosynthesis by Altering the Methionyl-tRNA Synthetase Activity of a Bacterial Expression Host13, Angewandte Chemie International Edition 39, 2148–2152. 4. Dieterich, D. C., Link, A. J., Graumann, J., Tirrell, D. A., and Schuman, E. M. (2006) Selective identification of newly synthesized proteins in mammalian cells using bioorthogonal noncanonical amino acid tagging (BONCAT), Proceedings of the National Academy of Sciences 103, 9482–9487. 5. Prescher, J. A., and Bertozzi, C. R. (2005) Chemistry in living systems, Nat. Chem Biol. 1, 13–21. 6. Waehler, R., Russell, S. J., and Curiel, D. T. (2007) Engineering targeted viral vectors for gene therapy, Nat. Rev. Genet. 8, 573–587. 7. Mathis, J. M., Stoff-Khalili, M. A., and Curiel, D. T. (2005) Oncolytic adenoviruses - selective retargeting to tumor cells, Oncogene 24, 7775–7791. 8. Henning, P., Lundgren, E., Carlsson, M., Frykholm, K., Johannisson, J., Magnusson, M. K.,

Tang, E., Franqueville, L., Hong, S. S., Lindholm, L., and Boulanger, P. (2006) Adenovirus type 5 fiber knob domain has a critical role in fiber protein synthesis and encapsidation, J. Gen. Virol. 87, 3151–3160. 9. Manuel, A. F. V. G., and Antoine, A. F. d. V. (2006) Adenovirus: from foe to friend, Reviews in Medical Virology 16, 167–186. 10. Mahal, L. K., Yarema, K. J., and Bertozzi, C. R. (1997) Engineering Chemical Reactivity on Cell Surfaces Through Oligosaccharide Biosynthesis, Science 276, 1125–1128. 11. Kim, E. J., Sampathkumar, S.-G., Jones, M. B., Rhee, J. K., Baskaran, G., Goon, S., and Yarema, K. J. (2004) Characterization of the Metabolic Flux and Apoptotic Effects of O-Hydroxyland N-Acyl-modified N-Acetylmannosamine Analogs in Jurkat Cells, Journal of Biological Chemistry 279, 18342–18352. 12. Agard, N. J., Baskin, J. M., Prescher, J. A., Lo, A., and Bertozzi, C. R. (2006) A Comparative Study of Bioorthogonal Reactions with Azides, ACS Chemical Biology 1, 644–648. 13. Dieterich, D. C., Lee, J. J., Link, A. J., Graumann, J., Tirrell, D. A., and Schuman, E. M. (2007) Labeling, detection and identification of newly synthesized proteomes with bioorthogonal non-canonical amino-acid tagging, Nat. Protocols 2, 532–540. 14. Laughlin, S. T., and Bertozzi, C. R. (2007) Metabolic labeling of glycans with azido sugars and subsequent glycan-profiling and visualization via Staudinger ligation, Nat. Protocols 2, 2930–2944.

Chapter 6 Preparation of Peptide and Other Biomolecular Conjugates Through Chemoselective Ligations Mathieu Galibert, Olivier Renaudet, Didier Boturyn, and Pascal Dumy Abstract The synthesis of molecular conjugates through chemoselective ligations represents a very convenient strategy to prepare complex macromolecules with diverse functional elements. Herein, we describe chemical methods based on the preparation of chemoselectively addressable peptides allowing successive oxime ligations and/or alkyne–azide cycloaddition (“click”) reactions of various biomolecules. This modular synthetic approach can be applied to a broad range of purposes. Key words: Peptide, Carbohydrate, Biomolecular conjugates, Chemoselective ligation, Oxime ligation, Alkyne–azide cycloaddition, Click chemistry

1. Introduction The use of efficient and chemoselective ligations is essential to construct biomolecular conjugates. The major barriers to the chemical ligation of biomolecules arise from their high molecular weight and/or from the incompatibility between the chemistries of peptides, nucleic acids and carbohydrates. To overcome these limitations, the fragment-coupling approach using chemoselective ligations has proved to be particularly useful to tailor proteins (1–3), oligonucleotide (4) or carbohydrate derivatives (5). The use of chemoselective reactions yielding disulfide, thioester, thiazolidine, hydrazone, or oxime bonds proved highly efficient for the preparation of relevant macromolecules such as synthetic vaccines (6, 7), synthetic proteins (8, 9), antiviral drugs (10), or anticancer agents (11). Previously, we successfully exploited oxime bond formation to prepare a diverse range Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_6, © Springer Science+Business Media, LLC 2011

67

68

Galibert et al.

of bioconjugates (12, 13); this reaction benefits from the high reactivity between aminooxy and carbonyl groups. Among other types of chemoselective reactions that have been described, the alkyne–azide cycloaddition (“click”) reaction (14, 15) is a powerful tool to prepare new peptidic derivatives through a triazole linkage (16). Very recently, we reported a novel strategy for the synthesis of biomolecular assemblies using orthogonal oxime bond formation and copper(I)-mediated alkyne–azide cycloaddition (CuAAC) reactions in either a stepwise or a onepot approach (17). We believe that this strategy could be a very convenient method for achieving the synthesis of highly sophisticated bioconjugate assemblies.

2. Materials 2.1. Reagents and Solutions

1. Fmoc-protected amino acids and resins. 2. (Benzotriazol-1-yloxy)tripyrrolidinophosphonium hexafluorophosphate (PyBOP). 3. 2,4,6-Trinitrobenzenesulfonic acid (TNBS) test reagents: DIPEA, DMF, and TNBS. 4. 4-Pentynoic acid succinimidyl ester. 5. Acetic acid (AcOH). 6. Acetone. 7. Acetonitrile (CH3CN). 8. Celite®. 9. Copper micro-powder. 10. Dichloromethane (CH2Cl2). 11. Diethyl ether (Et2O). 12. Diisopropylethylamine (DIPEA). 13. Dimethyldichlorosilane [Si(CH3)2Cl2]. 14. Dimethylformamide (DMF). 15. Dioxane. 16. Ethanol. 17. Ethyl acetate (EtOAc). 18. Ethyl N-hydroxyacetimidate. 19. HCl. 20. Iodoacetic acid. 21. Kaiser test reagents: Ninhydrin, phenol, pyridine, and potassium cyanide (KCN).

Preparation of Peptide and Other Biomolecular Conjugates

69

22. Magnesium sulfate (MgSO4) (anhydrous). 23. Methanol. 24. Methylhydrazine. 25. N-Hydroxyphthalimide. 26. N-Hydroxysuccinimide. 27. N,N ¢-Dicyclohexylcarbodiimide. 28. NaOH. 29. Pent-4-ynoic acid. 30. Pentane. 31. Piperidine. 32. Saturated aqueous solution of sodium chloride (brine). 33. Silica gel 60 (0.063–0.2 mm/70–230 mesh) (Merck). 34. Silica gel 60 F254 thin layer chromatography (TLC) plates (Merck). 35. Sodium bicarbonate (NaHCO3). 36. Sodium periodate (NaIO4). 37. Sodium sulfate. 38. Sulfuric acid. 39. tert-Butanol (tBuOH). 40. Triethylamine. 41. Trifluoroacetic acid (TFA). 42. Trifluoroethanol (TFE). 2.2. Equipment

1. Glass reaction vessels for manual solid-phase peptide synthesis (SPPS). 2. Handheld UV light source (see Note 1). 3. Nuclear magnetic resonance (NMR) spectrometer (see Note 2). 4. High performance liquid chromatography (HPLC) system equipped with analytical C18 columns (see Note 3). 5. Mass spectrometer (see Note 4).

3. Methods 3.1. Peptide and Functionalized Building Block Synthesis

Some examples of the amino acid building blocks synthesized in this section are shown in Fig. 1.

70

Galibert et al.

O

O O

OEt N

O

N O

1

O

O O

O

O

2

O

O

N

O O

N H

O

3

N

O

OEt N O

NH

O O

NH

O O

O OH

N H

O

O

4

O

O

AcOOAc O

O

A cO

NH

O O

OAc

OAc

AcO O

6

OH

OAc

O

O HO

OH

O

N O

O

HO OH N H

O

5

O

H N

O

OH

N H

7 OH

HO O

O

NH2

O OH

8

Fig. 1. Examples of amino acid and carbohydrate building blocks used in this work.

3.1.1. Solid-Phase Peptide Synthesis (see Note 5)

1. The linear peptides were assembled on 2-chlorotritylchloride resin using standard solid-phase peptide synthesis through an Fmoc/tBu strategy. 2. The coupling reactions were performed using, relative to the resin loading, 1.5–2 eq. of Fmoc-protected amino acid activated in situ with 1.5–2 eq. of PyBOP and 3–4 eq. of DIPEA in DMF (10 mL/g resin) for 30 min. 3. The resin was washed twice with DMF (10 mL/g resin) for 1 min and twice with CH2Cl2 (10 mL/g resin) for 1 min. 4. The completeness of the amino acid coupling reactions was checked by performing Kaiser and TNBS tests.

Preparation of Peptide and Other Biomolecular Conjugates

71

5. Na-Fmoc protecting groups were removed by treatment with piperidine/DMF (1:4, v/v) (10  mL/g resin) for 10  min, three times. 6. The resin was further washed five times with DMF (10 mL/g resin) for 1 min, and twice with CH2Cl2 (10 mL/g resin) for 1 min. 7. The completeness of the deprotection was checked by UV measurement (l = 299 nm, e = 7800 M−1 cm−1). 8. The cleavage and washing solutions were collected together and the volume of the combined solution was adjusted to a known value (V) with methanol. 9. Measurement of the absorbance of the solution obtained from step 8 at l = 299 nm enabled the amount of Fmoc protecting groups released from the cleavage to be calculated according to the Beer–Lambert relation. 3.1.2. Peptide Cleavage

1. Synthetic linear peptides were recovered directly upon repeated acid cleavage of the resins (Subheading  3.1.1) using TFE/ AcOH/CH2Cl2 (2:1:7, v/v/v) (10 mL/g resin) for 2 h. 2. The solvent was removed under reduced pressure and the residue dissolved in a minimum volume of CH2Cl2. 3. Ether was added to precipitate the peptide, then triturated and washed three times with ether to yield crude material without further purification.

3.1.3. Peptide Cyclization (see Note 6)

1. The linear peptide was dissolved in DMF (0.5 mM), and the pH of the resulting solution was adjusted to 8–9 by the addition of DIPEA. 2. The reagent PyBOP (1.2 eq.) was then added, and the solution was stirred at room temperature for 1 h (see Note 7). 3. The solvent was removed under reduced pressure, and the residue dissolved in a minimum volume of CH2Cl2. 4. Ether was added to precipitate the peptide, then triturated and washed three times with ether to yield crude material without further purification.

3.1.4. Synthesis of 4-Pentynoic Acid Succinimidyl Ester 1

1. Dissolve pent-4-ynoic acid (1 eq.) and N-hydroxysuccinimide (1 eq.) in EtOAc/dioxane (1:1, v/v, 1 mM) at 0°C. 2. Add N,N ¢-dicyclohexylcarbodiimide (1 eq.) in one portion at 0°C and stir at room temperature for 5 h. 3. Remove the formed dicyclohexylurea by filtration on a pad of Celite®. Evaporate the solvent to dryness under reduced pressure.

72

Galibert et al.

4. Dissolve the residue in EtOAc and wash the organic layer with 5% (w/v) aqueous NaHCO3 (two times), water (once), and brine (once) by using a separatory funnel. 5. Dry the organic layer over anhydrous MgSO4, filter, and evaporate the filtrate to dryness. 6. Precipitation from CH2Cl2/pentane (1:10, v/v) affords the product quantitatively as a white solid (see Note 8). 3.1.5. N-Hydroxysuccinimidyl 2-(1-Ethoxyethyli­ deneaminooxy)Acetate 2 (18)

1. Dissolve iodoacetic acid in water (1 M) at 0°C and add aqueous NaOH (1 eq., 40% w/w). 2. At room temperature, add ethyl N-hydroxyacetimidate (1.2 eq.) and aqueous NaOH (1.5 eq., 40% w/w) and stir at 80°C for 5 h. 3. Cool at room temperature and then add water (2 volumes). Wash the aqueous mixture with CH2Cl2 (three times). Bring the water phase to pH 2–3 with a 2 M HCl solution. Extract with EtOAc (four times) (see Note 9). 4. Wash the combined organic phases with brine (one time), dry over MgSO4, filter and evaporate to dryness to obtain the product as a colorless oil (73% yield). 5. React the product obtained in step 4 above with N-hydroxysuccinimide by following the procedure described previously in Subheading 3.1.4. 6. Precipitation from Et2O/pentane (1:10, v/v) affords the product quantitatively as a white solid (see Note 10).

3.1.6. Synthesis of Boc-Ser(tBu)-OSu 3

1. Follow the procedure described in Subheading 3.1.4 to react the commercial reagent Boc-Ser(tBu)-OH with N-hydroxysuccinimide. 2. Precipitation from Et2O/pentane (1:10, v/v) affords the product qualitatively as a white solid (see Note 11).

3.1.7. Fmoc-Lys[N-4Pentynoic Acid]-OH 4

1. Dissolve Fmoc-Lys-OH in DMF (1 mM) and adjust the pH to 8–9 by addition of DIPEA. 2. To the above mixture, add compound 1 (1  eq.) (obtained from Subheading 3.1.4) and stir the reaction at room temperature for 2 h. 3. Evaporate the solution to dryness under reduced pressure and dissolve the residue in EtOAc. 4. Wash the organic layer with concentrated citric acid solution (three times), water and brine. 5. Dry the organic layer over anhydrous MgSO4, filter, and evaporate the filtrate to dryness.

Preparation of Peptide and Other Biomolecular Conjugates

73

6. Precipitation from CH2Cl2/pentane (1:10, v/v) affords the product as a white solid (90% yield), which is used without further purification (see Note 12). 3.1.8. Fmoc-Lys[N-EeiAoa]-OH 5 (19)

1. Follow the procedure described in Subheading 3.1.7 to react compound 2 (1 eq.; obtained from Subheading 3.1.5) with Fmoc-Lys-OH. 2. Purify the crude product by column chromatography using CH2Cl2/EtOH (9:1, v/v) as eluent. 3. After evaporation, the desired compound is obtained as a white solid (71% yield) (see Note 13).

3.1.9. Fmoc-Lys[BocSer(tBu)]-OH 6

1. Follow the procedure described in Subheading 3.1.7 to react compound 3 (1 eq.; obtained from Subheading 3.1.6) with Fmoc-Lys-OH. 2. Precipitation from CH2Cl2/pentane (1:10, v/v) affords the product as a white solid (92% yield), which is used without further purification (see Note 14).

3.1.10. Glycosylation Between N-hydroxyphthalimide and a Glycosyl Fluoride: Synthesis of Phthalimido Lactosyl 7 (20) as an Example

1. Add N-hydroxyphthalimide (1 eq.) and triethylamine (1 eq.) to a solution of acetylated lactosyl fluoride in dry CH2Cl2 (1 M) (see Note 15). 2. Add the promoter BF3·Et2O (4 eq.) and stir the solution for 15 min at room temperature (see Note 16). 3. Add CH2Cl2 to the mixture and wash the organic layer with 10% (w/v) aqueous NaHCO3 (three times) and then with water. 4. Dry the organic layer under sodium sulfate, evaporate, and purify the resulting glycosylated derivative 7 by silica gel chromatography (CH2Cl2/EtOAc, 4:1, v/v). 5. Precipitation from CH2Cl2/pentane (1:10, v/v) affords 7 as a white powder (70% yield) (see Note 17).

3.1.11. Hydrazinolysis: Synthesis of Aminooxy Lactosyl 8 (20) as an Example

1. Dissolve the acetylated compound 7 (obtained from Subheading  3.1.10) in a solution of ethanol/methylhydrazine (1:1, v/v, 10 mL). 2. Evaporate the solvent under reduced pressure after stirring the solution overnight at room temperature. 3. Precipitation from MeOH/CH2Cl2 (1:10, v/v) affords the fully deprotected and pure b-aminooxylated lactose as a white amorphous powder (74% yield) (see Note 18).

3.2. Molecular Assembly of Bioconjugates

See (Fig. 2).

74

Galibert et al.

A: Ligation Oxime/Oxime O

O O

O K K

O

O K

A K

K

O

N

Sugar

O

N

O

O

G P

Sugar

O

O

Sugar

Sugar

P G

1)

Peptide

2)

Sugar

CHO

O O

N O K

G K

P

N O K

A K

K

P G

ONH2 O ONH2

O N

Peptide

CHO

2)

Peptide

de

Peptide

P G

N N N

pti

1)

O K

Pe

K

K

O

de

K

A

pti

G P

NH2O

O K

Pe

NH2O

O

de

de

pti

pti

Pe

Pe

B: Ligation Oxime/Click

O

N N N

N3

O N

O K

G P

NO

O

K

O O K

A K

K

P G

Fig. 2. General strategies for biomolecular assembly using oxime bond formation and copper(I)-mediated alkyne–azide cycloaddition (CuAAC; “click”) reactions.

3.2.1. Oxidative Cleavage of Serine Residues to Generate Aldehyde Groups

1. Dissolve the peptide containing serine residues in CH3CN/ H2O (2:1, v/v, 10 mM). 2. Add NaIO4 (10 eq. per serine) and stir the solution at room temperature for 30 min. 3. Purify the oxidized product (containing aldehyde groups) by reversed-phase HPLC (RP-HPLC) and lyophilize the combined fractions.

3.2.2. Oxime/Oxime Ligation (21) (Fig. 2a)

1. Dissolve the peptide containing aldehyde groups in CH3CN/ H2O/TFA (1:1:0.1, v/v/v, 10 mM). 2. Add peptide containing an oxyamine function (2 eq. per aldehyde site) and stir at 37°C overnight. 3. Quench the excess of oxyamine peptide with acetone and then oxidize the crude mixture with NaIO4 by following the procedure described in Subheading 3.2.1.

Preparation of Peptide and Other Biomolecular Conjugates

75

4. After semi-preparative HPLC, dissolve the aldehydic peptide in CH3CN/H2O/TFA (1:1:0.1, v/v/v, 10 mM). 5. Add the aminooxy sugar (2 eq. per aldehyde). 6. Stir reaction mixture at 37°C overnight and purify the final compound by semi-preparative RP-HPLC. 3.2.3. Oxime/CuAAC Ligation (17) (Fig. 2b)

1. Dissolve the peptide containing alkyne and oxyamine moieties in tBuOH/H2O/AcOH (50:45:5, v/v/v, 10 mM). 2. Add the aldehyde-containing peptide (or other aldehydecontaining biomolecule) (1.5 eq. per oxyamine site) and stir at room temperature for 1 h. 3. Adjust the pH to 8 using 10% (w/w) NaHCO3 solution, and then add the azide-containing peptide (or other azide-­ containing biomolecule) (1.5 eq. per alkyne site) and micropowder copper (5 eq.). Stir the reaction mixture overnight at room temperature. 4. Centrifuge the crude solution and purify the supernatant (containing the biomolecular conjugate) by semi-preparative RP-HPLC.

4. Notes 1. UV light is used for TLC spot visualization. Prior to visualization, the TLC plate is heated after treatment with a solution of 10% sulfuric acid in ethanol. 2. 1H and 13C NMR spectra were recorded on a Bruker AC300 spectrometer and the chemical shifts (d) are reported in parts per million (ppm). Spectra were referenced to the residual proton solvent peaks relative to the signal from CDCl3 (d 7.27 and 77.23 ppm for 1H- and 13C-NMR, respectively) or relative to the signal of D2O (d 4.79  ppm for 1H-NMR). Proton and carbon assignments were obtained from GCOSY and GHMQC experiments. The anomeric configuration of the carbohydrate samples was established by determination of the coupling constant (J) between H-1 and H-2. 3. The progress of the peptide synthesis reactions was monitored by reversed-phase HPLC using a Waters or equivalent HPLC system equipped with C18 columns. Analytical HPLC (Nucleosil 120 Å 3 mm C18 particles, 30 × 4.6 mm2) was performed at 1.3  mL/min and preparative HPLC (Delta-Pak 100 Å 15  mm C18 particles, 200 × 25  mm2) at 22  mL/min with UV monitoring at 214 and 250 nm using a linear A–B gradient (buffer A: 0.09% CF3CO2H in water; buffer B: 0.09% CF3CO2H in 90% acetonitrile).

76

Galibert et al.

4. Mass spectra were obtained by electron spray ionization (ESI-MS) on a VG Platform II or by chemical ionization (DCI-MS) on a Thermo Finnigan PolarisQ in the positive mode. 5. The assembly of all linear, protected peptides was performed manually in a glass reaction vessel fitted with a sintered glass frit or automatically on a peptide synthesizer (348 W Peptide Synthesizer, Advanced ChemTech). In manual SPPS, the glass reaction vessel allows elimination of excess reagents and solvents under compressed air. Before use, the glass reaction vessel was treated for 12  h (typically overnight) with (CH3)2SiCl2 as lubricant to prevent the resin beads from sticking to the glass inner wall during the synthesis reactions. The glass reaction vessel was then carefully washed with CH2Cl2 until the complete clearance of any trace amounts of acid was achieved. 6. The presence of a b-turn in the peptide structure facilitates the cyclization process. 7. After 5 min, it is necessary to control the pH. The pH should be adjusted to 8–9 by the addition of DIPEA. 8. Characterization of 4-pentynoic acid succinimidyl ester 1: 1H NMR (300 MHz, CDCl3): d = 2.03 (t, 1H, J = 2.4 Hz), 2.60 (td, 2H, J = 2.4, 7.0 Hz), 2.83 (s, 4H), 2.87 (t, 2H, J = 7.0 Hz); 13 C NMR (75 MHz, CDCl3) d = 14.1, 25.6, 30.3, 69.6, 80.3, 167.1, 169.0. 9. After each extraction, the pH is adjusted to 2–3 with a 2 M HCl solution. 10. Characterization of N-hydroxysuccinimidyl 2-(1-ethoxyethylideneaminooxy)acetate 2: 1H NMR (CDCl3, 300  MHz) d = 4.78 (s, 2H), 4.01 (q, 2H, J = 7.2 Hz), 2.84 (s, 4H), 1.98 (s, 3H), 1.28 (t, 3H, J = 7.2 Hz). 11. Characterization of Boc-Ser(tBu)-OSu 3: 1H NMR (300  MHz, CDCl3): d = 5.41 (d, 1H, J = 9.0  Hz), 4.78 (d, 1H, J = 9.0 Hz), 3.92 (m, 1H), 3.66 (m, 1H), 2.82 (s, 4H), 1.46 (s, 9H), 1.20 (s, 9H); ESI-MS (positive mode): calculated for C16H26N2O7: 358.2, found 359.3. 12. Characterization of Fmoc-Lys[N-4-Pentynoic acid]-OH 4: 1 H NMR (300 MHz, CDCl3) d = 7.75 (d, 2H, J = 7.4 Hz), 7.60 (d, 2H, J = 7.4 Hz), 7.38 (t, 2H, J = 7.4 Hz), 7.29 (t, 2H, J = 7.4  Hz), 6.0 (t, 1H, J = 5.6  Hz), 5.70 (d, 1H, J = 7.8  Hz), 4.38–4.36 (m, 3H), 4.12 (t, 1H, J = 6.9  Hz), 3.19 (m, 2H), 2.42 (t, 2H, J = 7.0 Hz), 2.30 (dt, 2H, J = 2.3, 7.0 Hz), 1.92 (t, 1H, J = 2.4 Hz), 1.96 (s, 3H), 1.81 (m,2H), 1.58 (m, 2H), 1.45 (m, 2H), 1.24 (t, 3H, J = 7.1 Hz); 13C NMR (CDCl3, 75  MHz) d = 173.9, 170.0, 156.1, 143.8,143.7, 140.6, 127.6, 127.0, 125.2, 120.0, 83.7, 71.2,

Preparation of Peptide and Other Biomolecular Conjugates

77

65.6, 53.7, 46.6, 38.2, 34.2, 30.4, 28.6, 23.0, 14.2; ESI-MS (positive mode): calculated for C26H28N2O5: 448.5, found 449.1. 13. Characterization of Fmoc-Lys[N-Eei-Aoa]-OH 5: 1H NMR (300  MHz, CDCl3) d = 7.75 (d, 2H, J = 7.4  Hz), 7.60 (d, 2H, J = 7.4  Hz), 7.38 (t, 2H, J = 7.4  Hz), 7.29 (t, 2H, J = 7.4 Hz), 6.50 (t, 1H, J = 5.6 Hz), 5.70 (d, 1H, J = 7.8 Hz), 4.38–4.36 (m, 5H), 4.20 (t, 1H, J = 6.9 Hz), 3.96 (q, 2H, J = 7.1 Hz), 3.34 (m, 2H), 1.96 (s, 3H), 1.81 (m, 2H), 1.58 (m, 2H), 1.45 (m, 2H), 1.24 (t, 3H, J = 7.1 Hz); 13C NMR (75  MHz, CDCl3) d = 174.7, 171.3, 164.4, 156.2, 143.7, 141.3, 127.7, 127.1, 125.1, 119.9, 72.6, 67.1, 62.8, 53.7, 47.2, 38.5, 31.7, 29.1, 22.1, 14.2, 13.9; ESI-MS (positive mode): calculated for C27H33N3O7: 511.2, found 512.1. 14. Characterization of Fmoc-Lys[Boc-Ser(tBu)]-OH 6: 1H NMR (300 MHz, CDCl3): d = 7.68 (d, 2H, J = 7.5 Hz), 7.55 (d, 2H, J = 7.5 Hz), 7.33–7.18 (m, 4H), 6.84 (broad s, 1H), 6.25 (broad d, 1H), 5.62 (broad s, 1H), 4.34–4.06 (m, 5H), 3.63 (m, 1H), 3.36 (m, 1H), 3.20–3.05 (m, 2H), 1.84 (m, 1H), 1.65 (m, 1H), 1.47–1.32 (m, 14H),1.09 (s, 9H); ESI-MS (positive mode): calculated for C33H45N3O8: 611.3, found 611.2. 15. The solution immediately turns orange after the addition of triethylamine. 16. The solution becomes colorless after the addition of the promoter. This decoloration indicates the completeness of the reaction. 17. Characterization of O-(2,3,4,6-tetra-O-acetyl-b-dgalactopyranosyl)-(1 → 4¢)-(2¢,3¢,6¢-tri-O-acetyl-b- d glucopyranosyl)-N-oxyphthalimide 7: 1H NMR (300 MHz, CDCl3): d = 7.86–7.74 (m, 4 H, Har.), 5.34 (dd, 1 H, J4¢,5¢ = 0.9 Hz, J3¢,4¢ = 3.3 Hz, H-4¢), 5.28–5.18 (m, 2 H, H-2, H-3), 5.14 (d, 1 H, J1,2 = 6.9  Hz, H-1), 5.10 (dd, 1 H, J1¢,2¢ = 7.8 Hz, J2¢,3¢ = 10.4 Hz, H-2¢), 4.95 (dd, 1 H, J3¢,4¢ = 3.3 Hz, J2¢,3¢ = 10.4 Hz, H-3¢), 4.53 (d, 1 H, J1¢,2¢ = 7.8 Hz, H-1¢), 4.42 (dd, 1 H, J5,6a = 2.2 Hz, J6a,6b = 12.1 Hz, H-6a), 4.15 (dd, 1 H, J5,6b = 5.8 Hz, J6a,6b = 12.1 Hz, H-6b), 4.11–4.07 (m, 3 H, H-4, H-6¢), 3.88 (td, 1 H, J4¢,5¢ = 0.9 Hz, J5¢,6¢ = 6.8 Hz, H-5¢), 3.78–3.73 (m, 1 H, H-5), 2.16, 2.13, 2.08, 2.06, 2.05, 2.03, 1.94 (7s, 21 H, 7OCOCH3); 13C NMR (75 MHz, CDCl3): d = 170.8 (C=OAc), 170.7 (C=OAc), 170.5 (C=OAc), 170.4 (C=OAc), 170.0 (C=OAc), 169.9 (C=OAc), 169.5 (C=OAc), 163.1 (C=OPht), 135.1 (CHar.), 129.1 (Car.), 124.2 (CHar.), 104.5 (C-1), 101.5 (C-1¢), 76.3 (C-4), 73.3, 73.1 (C-3, C-5), 71.4, 71.1 (C-3¢, C-5¢), 70.4 (C-2), 69.4 (C-2¢), 67.1 (C-4¢), 62.5 (C-6), 61.3 (C-6¢), 21.2 (OCOCH3), 21.1 (OCOCH3), 21.0 (OCOCH3), 20.9 (OCOCH3); ESI-HRMS

78

Galibert et al.

(positive mode): calculated for 804.19631 [M+Na]+, found: 804.19576.

C34H39NO20Na:

18. Characterization of O-(b-d-galactopyranosyl)-(1 → 4¢)-(b-dglucopyranosyl) oxyamine 8: 1H NMR (300  MHz, D2O): d = 4.62 (d, 1 H, J1¢,2¢ = 8.3  Hz, H-1¢), 4.47 (d, 1 H, J1,2 = 7.7  Hz, H-1), 4.03 (dd, 1 H, J5¢,6a¢ = 1.3  Hz, J6a¢,6b¢ = 11.6 Hz, H-6a¢), 3.95 (bd, 1 H, J3,4 = 3.1 Hz, H-4), 3.84 (dd, 1 H, J5¢,6b¢ = 4.6 Hz, J6a¢,6b¢ = 11.6 Hz, H-6b¢), 3.793.63 (m, 7 H, H-3, H-5, H-6, H-3¢, H-4¢, H-5¢), 3.66 (dd, 1 H, J1,2 = 7.7  Hz, J2,3 = 9.8  Hz, H-2), 3.38 (bt, 1 H, J2¢,3¢ = 8.4 Hz, H-2¢); 13C NMR (75 MHz, D2O): d = 105.2 (C-1¢), 103.3 (C-1), 78.6, 75.7, 75.1, 74.8, 72.8, 71.7 (C-2¢), 71.3 (C-2), 68.9 (C-4), 61.4 (C-6¢), 60.4 (C-6). ESIHRMS (positive mode): calculated for C12H23NO11Na: 380.11688 [M+Na]+, found: 380.11633.

Acknowledgements We thank the Université Joseph Fourier (UJF-Grenoble), the Centre National de la Recherche Scientifique (CNRS), and the NanoBio program (Grenoble) for providing support for this work. References 1. Dawson, P. E. and Kent S. B. H. (2000) Synthesis of native proteins by chemical ligation. Annu. Rev. Biochem. 69, 923–960. 2. Hang, H. C. and Bertozzi, C. R. (2001) Chemoselective approaches to glycoprotein assembly. Acc. Chem. Res. 34, 727–736. 3. Carrico, I. S. (2008) Chemoselective modification of proteins: hitting the target. Chem. Soc. Rev. 37, 1423–1431. 4. Venkatesan, N. and Kim, B. H. (2006) Peptide conjugates of oligonucleotides: synthesis and applications. Chem. Rev., 106, 3712–3761. 5. Peri, F. and Nicotra, F. (2004) Chemoselective ligation in glycochemistry. Chem. Commun., 623–627. 6. Verez-Bencomo, V., Fernández-Santana, V., Hardy, E., Toledo, M. E., Rodríguez, M. C., Heynngnezz, L., Rodriguez, A., Baly, A., Herrera, L., Izquierdo, M., Villar, A., Valdés, Y., Cosme, K., Deler, M. L., Montane, M., Garcia, E., Ramos, A., Aguilar, A., Medina, E., Toraño, G., Sosa, I., Hernandez, I., Martínez, R., Muzachio, A., Carmenates, A., Costa, L.,

Cardoso, F., Campa, C., Diaz, M. and Roy, R. (2004) A synthetic conjugate polysaccharide vaccine against haemophilus influenzae type b. Science 305, 522–525. 7. Zeng, W., Jackson, D. C., Murray, J., Rose, K. and Brown, L. E. (2000) Vaccine 18, 1031–1039. 8. Rose, K. (1994) Facile synthesis of homogeneous artificial proteins. J. Am. Chem. Soc. 116, 30–33. 9. Canne, L. E., Ferré-D’Amare, A. R., Burley, S. K. and Kent S. B. H. (1995) Total chemical synthesis of a unique transcription factorrelated protein: cMyc-Max. J. Am. Chem. Soc. 117, 2998–3007. 10. Naicker, K. P., Li, H., Heredia, A., Song H. and Wang, L.-X. (2004) Design and synthesis of aGal-conjugated peptide T20 as novel antiviral agent for HIV-immunotargeting. Org. Biomol. Chem. 2, 660–664. 11. Henry, M. D., Wen, S., Silva, M. D., Chandra, S., Milton, M. and Worland, P. J. (2004) A prostate-specific membrane antigen-targeted

Preparation of Peptide and Other Biomolecular Conjugates monoclonal antibody–chemotherapeutic conjugate designed for the treatment of prostate cancer. Cancer Res. 64, 7995–8001. 12. Garanger, E. Boturyn, D.; Renaudet, O. Defrancq, E. and Dumy, P. (2006) Chemoselectively addressable template: a valuable tool for the engineering of molecular conjugates. J. Org. Chem. 71, 2402–2410. 13. Renaudet, O., Boturyn, D. and Dumy, P. (2009) Biomolecular assembly by iterative oxime ligations. Bioorg. Med. Chem. Lett. 19, 3880–3883. 14. Tørnoe, C. W., Christensen, C. and Meldal, M. (2002) Peptidotriazoles on solid phase: [1,2,3]-triazoles by regiospecific copper(I)catalyzed 1,3-dipolar cycloadditions of terminal alkynes to azides. J. Org. Chem. 67, 3057–3064. 15. Rostovtsev, V. V., Green, L. G., Fokin, V. V. and Sharpless, K. B. (2002) A stepwise huisgen cycloaddition process: copper(I)-catalyzed regioselective ligation of azides and terminal alkynes. Angew. Chem. Int. Ed. 41, 2596–2599. 16. Angell, Y. L. and Burgess, K. (2007) Peptidomimetics via copper-catalyzed azide– alkyne cycloadditions. Chem. Soc. Rev. 36, 1674–1689.

79

17. Galibert, M., Dumy, P. and Boturyn, D. (2009) One-pot approach to well-defined biomolecular assemblies via orthogonal chemoselective ligations. Angew. Chem. Int. Ed. 48, 2576–2579. 18. Duléry, V., Renaudet, O. and Dumy, P. (2007) Ethoxyethylidene protecting group prevents N-overacylation in aminooxy peptide synthesis. Tetrahedron 63, 11952–11958. 19. Foillard, S., Olsten Rasmussen, M., Razkin, J., Boturyn, D. and Dumy, P. (2008) 1-ethoxyethylidene, a new group for the stepwise spps of aminooxyacetic acid-containing peptides. J. Org. Chem. 73, 983–991. 2 0. Renaudet, O. and Dumy, P. (2006) On-bead synthesis and binding assay of chemoselectively template-assembled multivalent neoglycopeptides. Org. Biomol. Chem. 4, 2628–2636. 21. Renaudet, O., Křenek, K., Bossu, I., Dumy, P., Kádek, A., Adámek, D., Vaněk, O., Kavan, D., Gažák, R., Šulc, M., Bezouška, K. and Křen, V. (2010) synthesis of multivalent glycoconjugates containing the immunoactive LELTE peptide: Effect of Glycosylation on cellular activation and natural killing by human peripheral blood mononuclear cells. J. Am. Chem. Soc. 132, 6800–6808.

Chapter 7 New Fluorescent Substrates of Microbial Transglutaminase and Its Application to Peptide Tag-Directed Covalent Protein Labeling Noriho Kamiya and Hiroki Abe Abstract Transglutaminase (TGase) is an enzyme that catalyzes the post-translational covalent cross-linking of Gln- and Lys-containing peptides and/or proteins according to its substrate specificity. We have recently designed a variety of Gln-donor fluorescent substrates of microbial transglutaminase (MTG) from Streptomyces mobaraensis and evaluated their potential use in MTG-mediated covalent protein labeling. The newly designed substrates are based on the relatively broad substrate recognition of MTG for the substitution of the N-terminal group of a conventional TGase substrate, benzyloxycarbonyl-l-glutaminylglycine (Z-QG). It is revealed that MTG is capable of accepting a diverse range of fluorophores in place of the N-terminal moiety of Z-QG when linked via a suitable linker. Here, we show the potential utility of a new fluorescent substrate for peptide tag-directed covalent protein labeling by employing fluorescein-4-isothiocyanate-b-Ala-QG as a model Gln-donor substrate for MTG. Key words: Fluorescent substrate peptide, Peptide tag, Site-specific protein modification, Transglutaminase

1. Introduction The site-specific labeling of proteins with small molecules has been widely employed for probing and/or utilizing protein functions both in vivo and in vitro. Among the current protein labeling strategies available, covalent protein labeling is superior in terms of its robustness, which is a key factor in a range of practical applications. In particular, the use of a short peptide tag that can be post-translationally modified by a variety of means is highly useful for directing site-specific protein modifications because the incorporation of such types of short tag sequences into the

Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_7, © Springer Science+Business Media, LLC 2011

81

82

Kamiya and Abe

N- and/or C termini of target proteins can typically be easily achieved by standard genetic manipulations. Moreover, the introduction of a short tag can help minimize perturbations of the intrinsic function of target proteins. To further the development of peptide tag-directed protein modification strategies, we have focused on the use of transglutaminase (TGase) – and more specifically, microbial transglutaminase (MTG) from Streptomyces mobaraensis. MTG catalyzes the post-translational covalent cross-linking of Gln- and Lyscontaining peptides and/or proteins and can also accept amineterminating small probe molecules in place of Lys-containing peptides, which is very useful for introducing a new chemical entity to a substrate protein (1, 2). The practical utility of MTG in biotechnological applications was validated by demonstrations of the site-specific conjugation of a native protein with synthetic polymers (3) and the synthesis of recombinant proteins tagged with genetically introduced substrate peptide tags (4, 5). The basic concept has now been extended to the first example of the enzymatic labeling of recombinant proteins with oligonucleotides chemically modified with a substrate peptide of MTG (6). Several other studies have also revealed that MTG-mediated protein modification can be used to achieve the site-specific immobilization of recombinant proteins to solid surfaces, as well (7–9). In this chapter, we first focus on the synthesis of fluorescein4-isothiocyanate (FITC)-b-Ala-QG (Fig. 1), a new fluorescent probe molecule that can be used in site-specific, covalent protein labeling reactions catalyzed by MTG (10). Following this, we describe how a substrate peptide of MTG consisting of six amino acids (MKHKGS, abbreviated hereafter as K6-tag) can be fused to the N terminus of Escherichia coli alkaline phosphatase (AP) through appropriate genetic modifications to generate N-terminal K6-tagged AP (NK6-AP). Finally, in the last part of this chapter, we describe the covalent labeling of NK6-AP, our model recombinant protein substrate, with the new fluorescent probe molecule FITC-b-Ala-QG.

Fig. 1. Molecular structure of a conventional MTG substrate (Z-QG) and two new substrate peptides (Flc-QG; FITC-bAla-QG) designed in this work.

New Fluorescent Substrates of Microbial Transglutaminase and Its Application

83

2. Materials A useful catalytic property of MTG, which can employ small probe molecules (e.g., cadaverine derivatives) containing a primary amine as substrates, allows the labeling of a specific Gln residue in native and recombinant proteins (11). However, there are a few cases where Gln-containing small fluorescent probes have been used for labeling a specific Lys residue in target proteins. For example, the pioneering work by Fuchsbauer and coworkers demonstrated that chemically labeling the C-terminal carboxylic group of benzyloxycarbonyl-l-glutaminylglycine (Z-QG) with monodansylcadaverine yielded a fluorolabeled Z-QG, and that this designed fluorescent substrate could be recognized by MTG (12). In subsequent reports, Z-QG and Z-QQPL derivatives labeled with fluorophores at the C termini were also prepared for MTG- and human tissue TGase-mediated labeling of IgG antibodies, respectively (13). Because all the existing fluorescent Glndonor substrates of MTG are comprised of the core structure, Z-QG, our laboratory’s first attempt to develop alternative donor substrates focused on simply altering the N-terminal Z moiety with a fluorophore. To this end, compound Flc-QG (Fig. 1) was prepared and applied to MTG-mediated conjugation with NK6-AP; however, no reaction was evident. On the other hand, it was revealed that the introduction of a b-Ala linker between the fluorophore and reactive Gln residue (e.g., FITC-b-Ala-QG; see Fig.  1) dramatically enhances the reactivity of the compound toward MTG. Interestingly, we further discovered that MTG can also accept a diverse range of fluorophores in place of the FITC moiety of FITC-b-Ala-QG (10). The following experimental procedures can be generalized for the use of different pairs of fluorescent substrates and recombinant proteins fused with a MTG-reactive Lys-containing peptide tag (see Note 1). 2.1. Synthesis of FITC-b-Ala-QG by Fmoc Solid-Phase Peptide Synthesis

1. Empty PD10 column (GE Healthcare). 2. H-Gly-Trt (2-Cl) resin, Fmoc-Gln(Trt)-OH, Fmoc-b-Ala-OH. 3. 2-(1H-Benzotriazole-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate (HBTU). 4. 1-Hydroxybenzotriazole (HOBt). 5. N,N-Dimethylformamide (DMF). 6. Dichloromethane. 7. Trifluoroacetic acid (TFA). 8. Piperidine (PPD) (20%, v/v) solution in DMF (PPD/DMF). 9. Triisopropylsilane (TIPS). 10. Fluorescein-4-isothiocyanate.

84

Kamiya and Abe

11. N,N-Diisopropylethylamine (DIEA). 12. tert-Butyl methyl ether. 13. a-Cyano-4-hydroxycinnamic acid (CHCA). 14. Reversed-phase high-performance liquid chromatography (RP-HPLC) system. 15. RP-HPLC Eluent A: 0.1% (v/v) TFA in water. 16. RP-HPLC Eluent B: Acetonitrile. 17. RP-HPLC column: Inertsil ODS-3, 10 × 250  mm (GL Science Inc.). 18. Matrix-assisted laser desorption/ionization time-of-flight mass spectrometer (MALDI-TOF MS) (Bruker Autoflex III). 19. MALDI-TOF matrix: 10  mg/mL CHCA in acetonitrile/ Milli-Q water/TFA (50:50:0.1, v/v). 2.2. Preparation of NK6-AP Recombinant Protein Modified with a MTG-Specific Peptide Tag

1. Plasmid pET22b(+) (Novagen). 2. Gene fragment encoding E. coli alkaline phosphatase (AP) (we used the phoa gene from plasmid pET20, a kind gift from Dr. Hiroshi Ueda of The University of Tokyo). 3. Restriction enzymes (ApaI, BamHI, HindIII, XhoI). 4. Pyrobest® DNA polymerase (Takara Bio Inc., Japan). 5. E. coli strain JM109. 6. E. coli strain BL21(DE3). 7. Isopropyl-1-thio-b-d-galactopyranoside (IPTG). 8. Ampicillin. 9. HisTrap™ HP chromatography column (GE Healthcare). 10. BCA protein assay kit (Pierce). 11. Luria-Bertani medium (LB medium): Dissolve 10  g bactotryptone, 5 g bacto-yeast extract, and 10 g NaCl in deionized water to a total volume of 1  L. Adjust the pH to 7.0 with sodium hydroxide and sterilize the medium by autoclaving for 20 min at 0.2 MPa before use. 12. Buffer A: 50 mM Tris–HCl (pH 7.4), 1 mM EDTA, and 20% (w/w) sucrose. 13. Buffer B: 20  mM Tris–HCl (pH 7.4), 500  mM NaCl, and 35 mM imidazole. 14. Buffer C: 20 mM Tris–HCl (pH 7.4), 500 mM NaCl, and 500 mM imidazole.

2.3. MTG-Mediated Labeling of NK6-AP with FITC-b-Ala-QG

1. MTG from Streptomyces mobaraensis (705 U/g, in a purified form provided by Ajinomoto Co. Ltd., Japan) (see Note 2). 2. ATTO AE-6500 mini-slab polyacrylamide gel electrophoresis (PAGE) system (ATTO Co., Japan).

New Fluorescent Substrates of Microbial Transglutaminase and Its Application

85

3. SDS–PAGE buffer D (4×, for separating gel preparation): 1.5 M Tris–HCl, pH 8.8. 4. SDS–PAGE buffer E (4×, for stacking gel preparation): 0.5 M Tris–HCl, pH 6.8. 5. SDS–PAGE buffer F (1× running buffer): Dissolve 3.03  g Tris, 14.4 g glycine, and 1 g SDS in deionized water to a total volume of 1 L. 6. 30% (w/v) Acrylamide/bisacrylamide stock solution (29:1 w/w with 3.3% crosslinking) (see Note 3). 7. 10% (w/v) SDS in water. 8. 10% (w/v) Ammonium persulfate (APS) in water. 9. N,N,N ¢,N ¢-Tetramethylethylenediamine (TEMED). 10. SDS–PAGE sample buffer (2×): 12% (v/v) 2-mercaptoethanol, 4% (w/v) SDS, and 20% (v/v) glycerol in 100 mM Tris– HCl, pH 6.8. 11. Molecular Imager FX Pro imaging system (Bio-Rad Inc., USA) equipped for fluorescence imaging. 12. Dimethyl sulfoxide (DMSO). 13. p-Nitrophenyl phosphate (pNPP). 14. Quick-Coomassie brilliant blue (CBB) protein staining kit (Wako Pure Chemical Industries, Ltd., Japan).

3. Methods 3.1. Synthesis of FITC-b-Ala-QG by SPPS

1. FITC-b-Ala-QG was manually synthesized by Fmoc solidphase peptide synthesis (SPPS) (14) using a PD10 column filled with H-Gly-Trt (2-Cl) resin (0.1 mmol). 2. The coupling reactions were conducted with 5 eq. (relative to the resin) of an Fmoc-amino acid activated in situ with 5 eq. of HBTU, 5  eq. of HOBt, and 10  eq. of DIEA in DMF (2.3 mL) for 2–3 h. Removal of the Fmoc-protecting group was achieved by treatment with PPD/DMF for 15 min. 3. The deprotection and cleavage of the target peptide from the resin was achieved by the addition of TFA/TIPS/H2O (95:2.5:2.5, v/v/v) for 1 h. 4. The mixture was filtered to remove the resin, and the filtrate was treated with tert-butyl methyl ether (three to five times the volume of the filtrate) to precipitate the cleaved peptide. 5. The collected solid was lyophilized to yield a yellow powder.

86

Kamiya and Abe

Fig. 2. Verification of the synthesis of FITC-b-Ala-QG by mass spectroscopic analysis.

6. The peptide was purified by RP-HPLC (UV monitoring at 493 nm; flow rate = 5 mL/min; 10–80% Eluent B in 30 min using a linear elution gradient). 7. The RP-HPLC fractions containing the peptide product were collected and lyophilized. 8. The synthesis of FITC-b-Ala-QG was confirmed by MALDITOF MS (detection mode: linear positive). Mass spectrum analysis: C31H29N5O10S; calculated exact mass 663.16; found m/z [M+H]+ 664.2 (Fig. 2). 3.2. Preparation of NK6-AP 3.2.1. Plasmid Vector Construction for NK6-AP Expression in E. coli

1. A DNA fragment encoding alkaline phosphatase (AP) was amplified by the polymerase chain reaction (PCR) using a plasmid encoding AP (pET20-AP) as the DNA template. The overall molecular cloning procedure is shown in Fig. 3 (see Note 4). The primer nucleotide sequences used for this PCR reaction were 5¢-GGG GGG ATC CAC CCC CAG AAA TGC CTG TTC TAG-3¢ (Primer 1) and 5¢-CCC CCA AGC TTC TCG AGT TTC AGC CCC AGA GCG GCT TTC ATG G-3¢ (Primer 2). The resultant gene fragment encoding

New Fluorescent Substrates of Microbial Transglutaminase and Its Application

87

Fig. 3. Cloning of the phoa gene encoding alkaline phosphatase (ALP) into plasmid pET22.

AP (with BamHI and XhoI sites) was digested with the restriction enzymes BamHI and XhoI, and cloned into a bacterial plasmid expression vector, pET22b(+), to generate the plasmid pET22-AP. 2. To attach the K6-peptide tag to AP, a second PCR reaction was conducted using pET22-AP as the DNA template. The overall procedure is shown in Fig. 4 (see Note 5). The primer nucleotide sequences used for this PCR reaction were 5¢GCC AGC CAG ACG CAG ACG CGC CGA GAC AGA-3¢ (Primer 3) and 5¢-GGT GGA TCC TTT ATG TTT CAT GGC CAT CGC CGG CTG GGC AG-3¢ (Primer 4). The resultant gene fragment encoding NK6-AP was again cloned into pET22-AP, but this time by ApaI/BamHI doubledigestion, to produce plasmid pET22-NK6-AP. 3.2.2. NK6-AP Protein Expression in E. coli

1. NK6-AP was expressed by transforming E. coli strain BL21(DE3) cells with the recombinant plasmid pET22-NK6AP obtained from Subheading  3.2.1 above. Transformants carrying the recombinant expression plasmid were grown in

88

Kamiya and Abe

Fig. 4. Construction of plasmid pET22-NK6-AP for the expression of NK6-AP in E. coli.

50  mL of LB medium supplemented with ampicillin (100  mg/L) at 37°C. The resulting overnight seed culture was used to inoculate 1 L of LB medium supplemented with ampicillin (100 mg/L), and the cells were grown at 37°C to an optical density (OD600) of 0.6. To induce the expression of NK6-AP protein, the incubation temperature was lowered to 27°C and IPTG was added to the culture to give a final concentration of 0.1 mM. The culture was then grown further for 24 h at 27°C. Following this, the cells were harvested by centrifugation, washed with Buffer A, and incubated in an ice-cold 5 mM MgCl2 aqueous solution (see Note 6). 2. The lysates were centrifuged (30,000 × g, 30 min) at 4°C and then purified with a Ni-NTA column (HisTrap™ HP) by using the hexahistidine-tag (His-tag) sequence attached to the C terminus of NK6-AP. After equilibrating the Ni-NTA column with Buffer B, the lysate sample was applied and the column was washed with Buffer B.

New Fluorescent Substrates of Microbial Transglutaminase and Its Application

89

3. The purification of NK6-AP was confirmed by SDS–PAGE (see Subheading  3.2.3) and N-terminal protein sequencing (see Note 7). 4. The protein concentration of the purified sample was determined by employing the BCA assay kit according to the manufacturer’s instructions (using bovine serum albumin as the standard). 5. The catalytic activity of NK6-AP was measured by using pNPP as the substrate: To 1 mL of 1 M Tris–HCl buffer (pH 8.0), the hydrolysis of pNPP (1  mM) was initiated by the addition of NK6-AP (30 nM) at 25°C (see Note 8). 3.2.3. SDS–PAGE Analysis of Protein Samples

1. The glass plates for the analysis were scrubbed clean with 20% ethanol prior to use. 2. A monomer solution for a 10% (w/v) acrylamide separating gel was prepared by combining 2.67 mL of 30% (w/v) acrylamide/bisacrylamide solution, 2 mL of Buffer D, 3.25 mL of water, 80 mL of 10% (w/v) SDS, 6 mL of TEMED, and 27 mL of 10% (w/v) APS, followed by gentle but thorough mixing. The mixture was poured between the glass plates, leaving space for a stacking gel, and overlaid with water-saturated n-butanol. The separating gel was allowed to polymerize for about 1 h. 3. The n-butanol was poured off and the top of the gel was rinsed with deionized water. 4. A monomer solution for a stacking gel was prepared by combining 0.4 mL of 30% (w/v) acrylamide/bisacrylamide solution, 0.625 mL Buffer E, 1.44 mL water, 25 mL 10% (w/v) SDS, 4 mL TEMED, and 8.3 mL 10% (w/v) APS, followed by gentle but thorough mixing. The mixture was poured on top of the separating gel, the comb was inserted, and the stacking gel was overlaid with water-saturated n-butanol. The stacking gel was allowed to polymerize for 30 min. 5. Once the gel polymerization had completed, the comb was carefully removed and the wells were rinsed with deionized water. 6. Samples for SDS–PAGE analysis were prepared by mixing 10 mL of the sample with an equivalent volume of 2× sample buffer, followed by heat-treatment at 95°C for 15 min. Before applying to the wells of the gel, the samples were incubated for 30 min at 37°C. 7. Running buffer was added to the upper and lower chambers of the electrophoresis unit, and the samples (10  mL) were loaded into the wells. One well was loaded with prestained protein molecular weight markers (see Note 9).

90

Kamiya and Abe

8. After completing the assembly of the electrophoresis unit and connecting to a power supply, the gel was run at 250 V for 100 min. 9. The gel was stained with CBB using the Quick-CBB protein staining kit according to the manufacturer’s instructions. 3.3. MTG-Mediated Labeling of NK6-AP with FITC-b-Ala-QG

1. A stock solution (20 mM) of the fluorescent substrate FITCb-Ala-QG was prepared in DMSO. 2. The labeling reaction mixture was comprised of NK6-AP protein (0.5 mg/ml) and FITC-b-Ala-QG substrate (1 mM) in 100 mM Tris–HCl buffer (pH 8.0) containing of 5% (v/v) DMSO (see Note 10). 3. The labeling reaction was initiated by the addition of MTG (0.1 U/ml), and then the mixture was incubated at 4°C for 6 h (see Note 11). 4. To follow the time course of the protein labeling reaction, the labeling reaction products were analyzed by SDS–PAGE (see Subheading  3.2.3) (Fig.  5). A small aliquot of the reaction mixture was removed at periodic time intervals and mixed with SDS–PAGE 2× sample buffer to terminate the MTG reaction (see Note 12). 5. After completing the electrophoresis run, the in-gel fluorescence from the FITC-b-Ala-QG label appended to NK6-AP should be analyzed before staining the gel with CBB. The progress of the reaction can be followed by the increase in the fluorescence of the protein bands in the fluorescence image of the SDS–PAGE gel (Fig. 6) (see Note 9). In this work, the fluorescence intensity of the SDS–PAGE gel was measured at

Fig. 5. SDS–PAGE analysis of the MTG-mediated labeling of NK6-AP and wild-type AP with FITC-b-Ala-QG. Left : Gel image after staining with Coomassie brilliant blue stain. Right : Corresponding fluorescence image of the gel before CBB staining.

New Fluorescent Substrates of Microbial Transglutaminase and Its Application

91

Fig. 6. SDS–PAGE analysis of the time course of the MTG-mediated labeling of NK6-AP with FITC-b-Ala-QG. The progress of the labeling reaction was followed by monitoring changes in the fluorescence signal intensities derived from labeled protein fractions. Top: Gel image after staining with Coomassie brilliant blue stain. Bottom: Corresponding fluorescence image of the gel before CBB staining.

room temperature with a fluoroimager system using excitation at 488 nm and a 530 ± 15 nm band-pass emission filter. A quantitative analysis of the measured signal intensities was conducted by using the Quantity One® software provided by the fluoroimager instrument manufacturer. For purposes of following the time course of the protein labeling reaction, the maximum fluorescence signal intensity measured in the gel was defined as 100%. 6. Under the described experimental conditions, the labeling of NK6-AP was observed to be nearly complete within 1 h. To ensure complete labeling, however, we recommend that the labeling reactions should be incubated for 6 h (see Note 13).

4. Notes 1. The simple and straightforward approaches described in this chapter are potentially applicable to labeling any type of recombinant protein; however, it should be noted that nonspecific modification may occur depending on the tertiary structure of substrate proteins. Based on our own investigations with several recombinant proteins (e.g., bacterial alkaline phosphatase (6, 8, 10), dihydrofolate reductase (5), enhanced green fluorescent protein (4), glutathione S-transferase (15), and cytochrome P450 protein components from Pseudomonas putida (16)), it has been revealed that, in general, unstable proteins tend to be recognized and crosslinked

92

Kamiya and Abe

more easily by MTG. However, even though a protein of interest may contain a structurally disordered region, the protein may not be subjected to MTG recognition if there is no Gln or Lys residue present and the overall protein scaffold is stable enough. A recent comprehensive review by Fontana et al. covers the important physical and structural factors of substrate proteins that are susceptible to TGase-mediated site-specific post-translational protein modification (17). 2. MTG can be commercially obtained and purified according to the reported protocol (18). The catalytic activity of MTG was measured by using the colorimetric hydroxamate procedure with Z-QG as described previously in ref.19. One unit (U) of activity is defined as the amount of enzyme that catalyzes the formation of 1  mmol of hydroxamate per minute using l-glutamic acid g-monohydroxamate as the standard. 3. Caution: Be careful in handling the acrylamide/bis solution since acrylamide monomer is a potent neurotoxin. 4. To introduce the K6-tag just before the pelB coding sequence of pET22b(+) in the subsequent step, the BamHI site in the plasmid was used to provide the Gly-Ser-encoding gene sequence (Fig. 4). 5. The PCR reaction with Primer 4 makes it possible to introduce the K6-tag via a BamHI site on the primer. Primer 3 was designed by overlapping the single ApaI site in the plasmid, which makes it easier to purify the PCR product by agarose gel electrophoresis. 6. The plasmid pET22-NK6-AP was designed to produce NK6-AP in the periplasmic space of the host E. coli cells. The enzyme product is isolated from the periplasm by osmotic shock with treatment of Buffer A, and subsequent extraction using an aqueous MgCl2 solution. If the protein yield is low, then cell disruption by gentle sonication may also be applied. 7. To confirm the expression of the protein of interest, the N-terminal sequence should be checked. In the case of purified NK6-AP, the N-terminal amino acid sequence was clearly identified as being MKHKG, suggesting that intrinsic alkaline phosphatase from the E. coli host did not contaminate the purified protein preparation. 8. E. coli alkaline phosphatase exists in a dimeric quaternary structure. For the alkaline phosphatase activity assays, the molar concentration of NK6-AP was calculated based on the monomeric unit. 9. When following the time course of the labeling reaction by the increase in the fluorescence intensity derived from the protein band in the gel, the gel should be analyzed using a

New Fluorescent Substrates of Microbial Transglutaminase and Its Application

93

fluoroimager instrument before staining with any dyes. To visualize the protein molecular weight markers in the analysis, one can use commercial fluorescent molecular weight markers (e.g., DyLight fluorescent protein molecular weight markers from Pierce). 10. Although DMSO was employed in this work as the solvent for preparing stock solutions to increase the solubility of the synthetic substrates, the use of DMSO is not always required. In the case of FITC-b-Ala-QG, the fluorescent substrate can be dissolved in aqueous solutions up to a concentration of 1 mM in 100 mM Tris–HCl buffer, pH 8. 11. To shorten the overall reaction time, the protein labeling reaction can be conducted at higher temperature (e.g., 37°C) and/ or by using a higher MTG concentration (e.g., 1 U/mL). 12. MTG-catalyzed reactions can also be terminated by the addition of N-ethylmaleimide (NEM) to the reaction mixture under approximately neutral pH conditions. The NEM working concentration should be adjusted according to the MTG concentration used in the labeling reaction. In our case, we employed NEM at a final concentration of 1 mM. 13. For the direct identification of the labeling site of NK6-AP, one can conduct peptide mapping analyses with MALDITOF MS (4, 10). In analyzing MTG-mediated protein labeling reactions with a Gln-donor substrate (e.g., Z-QG), digestion of the labeled proteins with a protease whose substrate recognition site includes Lys residues is very useful because this provides different MS patterns before and after MTG-mediated protein labeling with a specific Lys residue on the target protein. In this work, we found that the second Lys residue in the K6-tag (MKHKGS, underlined) was specifically labeled (10). For the control experiments, the MTGmediated labeling of a synthetic N-terminal peptide, MKHKGSTPEMPVLENR, with a Gln-donor substrate should be conducted, which will provide a clear indication of the actual protein labeling site.

Acknowledgments We are grateful to Ajinomoto Co., Inc. (Japan) for providing us with the MTG sample. This work was supported by the Research for Promoting Technological Seeds from the Japan Science and Technology Agency (JST) of Japan, and also in part by a Grantin-Aid for the Global COE Program, “Science for Future Molecular Systems” from the Ministry of Education, Culture, Sports, Science and Technology (MEXT) of Japan.

94

Kamiya and Abe

References 1. O’Hare, H. M., Johnsson, K. and Gautier, A. (2007) Chemical probes shed light on protein function. Curr. Opin. Struct. Biol. 17, 488–494. 2. Sunbul, M. and Yin, J. (2009) Site specific protein labeling by enzymatic posttranslational modification Org. Biomol. Chem. 7, 3361–3371. 3. Sato, H., Hayashi, E., Yamada, N., Yatagai, M., Takahara, Y. (2001) Further studies on the site-specific protein modification by microbial transglutaminase. Bioconjugate Chem. 12, 701–710. 4. Kamiya, N., Tanaka, T., Suzuki, T., Takazawa, T., Takeda, S., Watanabe, K. and Nagamune, T. (2003) S-peptide as a potent peptidyl linker for protein crosslinking by microbial transglutaminase from Streptomyces mobaraensis Bioconjugate Chem. 14, 351–357. 5. Tanaka, T., Kamiya, N., Nagamune, T. (2005) N-terminal glycine-specific protein conjugation catalyzed by microbial transglutaminase FEBS Lett. 579, 2092–2096. 6. Tominaga, J., Kemori, Y., Tanaka, Y., Maruyama, T., Kamiya, N. and Goto, M. (2007) An enzymatic method for site-specific labeling of recombinant proteins with oligonucleotides Chem Commun. 401–403. 7. Josten, A., Meusel, M., Spencer, F., and Haalk, L. (1999) Enzyme immobilization via microbial transglutaminase: a method for the generation of stable sencing surface. J. Mol. Catal. B. Enzymol. 7, 58–66. 8. Tominaga, J., Kamiya, N., Doi, S., Ichinose, H., Maruyama, T. and Goto, M. (2005) Design of a specific peptide tag that affords covalent and site-specific enzyme immobilization catalyzed by microbial transglutaminase Biomacromolecules 6, 2299–2304. 9. Wong, L. S., Khan, F. and Micklefield, J. (2009) Selective Covalent Protein Immobilization: Strategies and Applications Chem. Rev. 109, 4025–4053. 10. Kamiya, N., Abe, H., Goto, M., Tsuji, Y. and Jikuya, H. (2009) Fluorescent substrates for covalent protein labeling catalyzed by microbial transglutaminase Org. Biomol. Chem. 7, 3407–3412.

11. Meusel, M. (2004) Synthesis of hapten-protein conjugates using microbial transglutaminase in Methods in Molecular Biology, Bioconjugation Protocols: Strategies and Methods 283, 109–123. 12. Pasternack, R., Laurent, H.-P., Rüth, T., Kaiser, A., Schön, N. and Fuchsbauer, H.-L. (1997) A fluorescent substrate of transglutaminase for detection and characterization of glutamine acceptor compounds Anal. Biochem. 249, 54–60. 13. Mindt, T. L., Jungi, V., Wyss, S., Friedli, A., Pla, G., Novak-Hofer, I., Grüngerg, J. and Schibli, R. (2008) Modification of different IgG1 antibodies via glutamine and lysine using bacterial and human tissue transglutaminase Bioconjugate Chem., 19, 271–278. 14. Garanger, E., Aikawa, E., Reynolds, F., Weissleder, R. and Josephson, L. (2008) Simplified syntheses of complex multifunctional nanomaterials Chem. Commun. 4792–4794. 15. Tanaka, Y., Tsuruda, Y., Nishi, M., Kamiya, N. and Goto, M. (2007) Exploring enzymatic catalysis at a solid surface: a case study with transglutaminase-mediated protein immobilization Org. Biomol. Chem., 5, 1764–1770. 16. Hirakawa, H., Kamiya, N., Tanaka, T., Nagamune, T. (2005) Intramolecular electron transfer in a cytochrome P450cam system with a site-specific branched structure Protein Eng. Des. Sel., 20, 453–459. 17. Fontana, A., Spolaore, B., Mero, A. and Veronese, F. M. (2008) Site-specific modification and PEGylation of pharmaceutical proteins mediated by transglutaminase Adv. Drug. Deliv. Rev., 60, 13–28. 18. Ando, H., Adachi, M., Umeda, K., Matsuura, A., Nonaka, M., Uchio, R., Tanaka, H. and Motoki, M. (1989) Purification and characteristics of a novel transglutaminase derived from microorganisms. Agric. Biol. Chem., 53, 2613–2617. 19. Folk, J. E. and Cole, P. W. (1965) Structural requirements of specific substrates for guinea pig liver transglutaminase J. Biol. Chem. 240, 2951–2960.

Chapter 8 Covalent Conjugation of Poly(Ethylene Glycol) to Proteins and Peptides: Strategies and Methods Anna Mero, Chiara Clementi, Francesco M. Veronese, and Gianfranco Pasut Abstract PEGylation, the covalent linking of PEG chains, has become the leading drug delivery approach for proteins. This technique initiated its first steps almost 40 years ago, and since then, a variety of methods and strategies for protein–polymer coupling have been devised. PEGylation can give a number of relevant advantages to the conjugated protein, such as an important in vivo half-life prolongation, a reduction or an abolishment of immunogenicity, and a reduction of aggregation. Furthermore, the technique has demonstrated a great degree of versatility and efficacy – not only PEG–protein conjugates have reached the commercial marketplace (with nine types of derivatives), but a PEG-aptamer and PEGylated liposomes are now also available. Most of this success is due to the development of several PEGylation strategies and to the large selection of PEGylating agents presently at hand for researchers. Nevertheless, this technique still requires a certain level of familiarity and knowledge in order to achieve a positive outcome for a PEGylation project. To draw general guidelines for conducting PEGylation studies is not always easy or even possible because such experiments often require case-by-case optimization. On the other hand, several common methods can be used as starting examples for the development of tailormade coupling conditions. Therefore, this chapter aims to provide a basic introduction to a wide range of PEGylation procedures for those researchers who may not be familiar with this field. Key words: Poly(ethylene glycol) (PEG), PEGylation, PEG–protein conjugate, Protein modification

1. Introduction PEGylation, the covalent attachment of PEG moieties to a therapeutic agent, was first reported by Abuchowski et al. in the 1970s (1) who demonstrated the usefulness of the strategy to improve the therapeutic value of proteins and peptides. Since then, PEGylation has been long studied in the literature and several conjugates have already reached the marketplace. The most

Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_8, © Springer Science+Business Media, LLC 2011

95

96

Mero et al.

prominent effect of PEGylation is a prolonged circulation time for therapeutic conjugates due to a decreased rate of kidney clearance and a reduction of proteolysis. Together, these advantages have lead to lower doses of administration and increased compliance by patients (2, 3). PEG is synthesized by the anionic ring opening polymerization of ethylene oxide initiated by nucleophilic attack of a hydroxide or a methoxide ion to the epoxide ring. The polymer can be obtained with a low polydispersity and a low content of impurities. Several derivatives of PEG that vary in molecular weight (300 Da to 40 kDa), structure (linear or branched), and reactive moiety are currently available from a number of different companies (e.g., Iris biotech, Laysan Bio, NOF, etc.). The choice of an appropriate PEG derivative to use will depend on the particular features of a given protein of interest, such as the primary sequence, chemical reactivity of available functional groups, molecular weight, biological activity, and function. In this chapter, we outline several strategies for carrying out the PEGylation of proteins and peptides, and also present a number of methods for characterizing and purifying PEG–protein conjugates that have been used in our own laboratory or published in the literature. 1.1. Characterization of PEGylating Agents

The development of an efficient and successful PEGylation study requires reliable methods for analyzing PEG reagents and PEG– protein conjugates at various stages of the process. In fact, PEG reagents contribute to a substantial portion of the manufacturing costs associated with PEGylated proteins, and their purity impacts the conjugation efficiency and overall product quality. The determination of the exact molecular weight, polydispersity index, presence of reactive and nonreactive impurities, degree of activation, and the presence of PEG dimers in the raw material must all be carefully evaluated.

1.1.1. NMR Spectroscopy of PEGylating Agents

To evaluate the degree of activation and to detect the presence of impurities (4), PEGs can be analyzed by both 1H- and 13C-NMR spectroscopy (Fig. 1). This technique requires a few milligrams of polymer and a suitable deuterated solvent such as dimethyl sulfoxide (DMSO-d6), chloroform (CDCl3), or water (D2O). Usually, CDCl3 or DMSO are preferred for the analysis of those activated PEGs that can be hydrolyzed in D2O.

1.1.2. Determination of PEG Diol Impurities in Methoxy-PEG Batches

The most utilized PEG for protein modification is methoxy-PEG (mPEG), where only one terminal end of the polymer can be activated while the other is capped with a methoxy group, preventing undesired intra- or intermolecular crosslinking (5). During the synthesis of mPEG, the anionic polymerization of ethylene oxide initiated by CH3O– eventually contains PEG diol

Covalent Conjugation of Poly(Ethylene Glycol) to Proteins and Peptides

97

Fig. 1. NMR spectrum of Boc-PEG-NHS in CDCl3.

secondary by-products due to the presence of trace amounts of water. In fact, OH− formed by water present in the reactor can itself initiate the polymerization process, yielding a chain that can grow at both ends. This process leads to the formation of diol PEG impurities (HO–PEG–OH) with the peculiarity of a double-MW value with respect to the MW of the mPEG batch. Size-exclusion chromatography (SEC) analysis, as described in Subheading 1.3.3 below, is suitable for the determination of such PEG diol species in mPEG batches. 1.1.3. Activation Degree of Amino-Reactive PEGylating Agents

The degree of activation of most amino-reactive PEGs can be determined by using a spectroscopic assay based on the so-called Glycyl-Glycine (Gly-gly) test. In this procedure, the degree of PEG activation is determined by reacting an equimolar amount of Gly-Gly with the activated PEG, followed by performing a Snyder and Sobocinsky colorimetric assay of the unreacted dipeptide (6). This assay uses 2,4,6-trinitrobenzenesulfonic acid (TNBS), which reacts stoichiometrically with primary amino groups in an alkaline medium to give a trinitrophenyl derivative absorbing at 420 nm (Fig. 2).

1.1.4. Activation Degree of Thiol-Reactive PEGylating Agents

To determine the degree of activation of thiol-reactive PEG reagents (e.g., PEG-OPSS, PEG-Mal), the polymer is mixed with an equimolar amount of a suitable molecule containing a free thiol (e.g., cysteine (Cys) or glutathione (GSH)). The presence of unreacted thiol groups is determined by the Ellman assay,

98

Mero et al.

Fig. 2. Reaction between 2,4,6-trinitrobenzensulfonic acid and a free primary amine group.

Fig. 3. Reaction between 5,5¢-dithiobis(2-nitrobenzoic acid) and a free thiol group.

Table 1 Hydrolysis half-lives of PEG-NHS species at pH 8, 25°C PEG-NHS ester

Active groups structure

Half-life (min)

PEG–(CH2)4–CO2–NHS

Succinimidyl valerate (SVA)

33.6

PEG–O–CO2–NHS

Succinimidyl carbonate (SC)

20.4

PEG–O2C–(CH2)3–CO2–NHS

Succinimidyl glutarate (SG)

17.6

PEG–O2C–(CH2)–CO2–NHS

Succinimidyl succinate (SS)

9.8

PEG–O–CH2–CO2–NHS

Succinimidyl carboxymethyl (SCM)

0.75

PEG–O–(CH2)–CO2–NHS

Succinimidyl propionate (SPA)

16.5

which is based on the use of 5,5¢-dithiobis (2-nitrobenzoic acid) (DTNB). In this assay, free thiols react with DTNB at neutral pH to give 2-nitro-5-thiobenzoic acid (TNB), a derivative absorbing at 412 nm (Fig. 3) (7). 1.1.5. Reactivity of NHS-Activated PEGs

The reactivity of N-hydroxysuccinimide (NHS)-activated PEGs can be easily evaluated by following the increase in UV absorbance at 260 nm due to the release of the NHS group when the reaction is performed at room temperature in borate buffer. The reactivity of several representative PEG-NHS species, expressed in terms of the hydrolysis rate, is reported in Table 1 (8). It is worth noting that the hydrolysis rate of such acylating

Covalent Conjugation of Poly(Ethylene Glycol) to Proteins and Peptides

99

PEGs depends upon the nature of the chemical group adjacent to the active ester. Another point to be highlighted is that in all cases, the aminolysis rate is always higher than the hydrolysis rate due to the higher nucleophilicity of free amino groups. 1.1.6. Mass Spectrometric Analyses of PEGylating Agents

PEG molecular weight determination can be rather difficult because of polymer polydispersity. For such determinations, matrix-assisted laser-desorption ionization (MALDI) mass spectrometry (MS) has been the technique most often employed because it produces mainly monocharged ions, thus generating mass spectra of low complexity (9–11). The specific operating conditions will depend upon the particular MALDI-TOF instrument employed, but using an acceleration voltage of 20 kV with linear detection is a suitable general method. The MALDI-TOF mass spectrum of a PEG sample typically shows a monomodal distribution of MW values with the main mass signals spaced apart by Dm/z = 44, in agreement with the monomer unit mass of the oxyethylene unit (44.053 g/mol) (Fig. 4) (12). Electrospray ionization mass spectrometry (ESI-MS), a method widely used for protein characterization, has a strong tendency to form multiply charged ions of the samples, thus hampering the analysis of polydisperse PEG polymers with molecular masses above a few kilodaltons.

1.1.7. Determination of Reactive, Low-Molecular Weight Impurities in PEG Batches

The impurities found in raw PEG preparations can come from either the process of synthesis of the polymer or the activation reaction, while others are formed by the degradation of the polymer (e.g., oxidation) or by cleavage of the chain itself. PEG-aldehyde, for example, is easily oxidized in air and low-molecular weight aldehydic impurities can often be found in the raw product.

Fig. 4. MALDI-TOF mass spectrum of PEG 6,000 Da.

100

Mero et al.

MS analysis, reversed-phase high performance liquid chromatography (RP-HPLC), or NMR spectroscopy are all useful methods to detect the presence of these substances (13). 1.2. Protein PEGylation

The PEGylation of proteins can be achieved either by a direct chemical reaction between an amino acid residue and a suitable PEGylating reagent, or by an enzyme-catalyzed linkage. Brief descriptions of several strategies for conjugating PEG molecules to proteins and peptides are presented below.

1.2.1. Random PEGylation at Free Amino Groups

The primary amino groups of proteins are good nucleophiles, and as such are exploited most frequently for PEG coupling by reaction in mildly basic media (pH 8.0–9.5). Lysines, commonly located on protein surfaces, are relatively abundant and easily accessible to reactive PEG reagents (e.g., activated PEGcarboxylates and PEG-carbonates; see Fig.  5a, b). Random PEGylation at these amino acid residues typically yields mixtures of different isomers and different degrees of modification. To a lesser extent, such reactive PEGylating agents can also target other types of nucleophilic groups found in proteins such as the side chains of serine, threonine, tyrosine, and histidine (14, 15). In this case, it is possible to cleave any unstable bonds by treating the resultant conjugate mixture with hydroxylamine, thus improving the homogeneity of the final PEGylated product (16).

Fig. 5. Coupling of protein amines with activated PEG-carbonates (a1, a2), PEG-carboxylates (b) and PEG-aldehyde (c).

Covalent Conjugation of Poly(Ethylene Glycol) to Proteins and Peptides

101

1.2.2. Protein N-Terminal PEGylation

In order to guarantee a higher degree of homogeneity of the product, it is possible to direct the PEG coupling reaction to take place only at the N terminus of a protein. This selectivity is possible by taking advantage of the different pKa values between the e-NH2 of lysine and the a-NH2 of the N terminus. By lowering the pH of the reaction mixture to ~5–6, all the e-amines in a protein will tend to be protonated whereas the a-NH2 group will still be partially present as a free base available for coupling with activated PEG molecules. This method generally gives optimal results when less reactive PEG-aldehydes are used. In these reactions, an unstable Schiff base is initially obtained, which is in turn reduced to a stable secondary amine (Fig. 5c). Several papers describe the modification of primary amines with PEG-acetaldehyde and later with the more stable PEG-propionaldehyde (17, 18). This conjugation method has been successfully exploited for the preparation of several PEG–protein conjugates; among these Neulasta®, an N-terminal mono-PEGylated granulocyte colony-stimulating factor (G-CSF) has demonstrated therapeutic and marketing success (19).

1.2.3. Thiol PEGylation

Cysteine residues are valuable targets for achieving the site-specific modification of proteins or peptides, and are present in the free form at a relatively low natural abundance level compared to the oxidized cystine species. Nevertheless, cysteines – if present – are often found partially or fully buried within the structure of proteins with limited accessibility to chemical reagents (20). Under appropriate conditions, cysteine residues can be modified selectively, rapidly, quantitatively, and either in a reversible or irreversible fashion (21). Furthermore, thanks to its relatively facile coupling chemistry, there are several examples of the insertion of cysteines by genetic engineering at desired positions in a protein sequence for site-specific conjugation (22). PEG-maleimide (PEG-Mal), PEG-vinyl sulfone (PEG-VS), or PEG-iodo acetamide (PEG-IA) derivatives have been used to obtain stable, irreversible thioether bonds between polymers and proteins (Fig. 6). PEG-orthopyridyl disulfide (PEG-OPSS)

Fig. 6. Thiol-reactive PEGs (a) and conjugation of PEG to a free thiol group (b).

102

Mero et al.

is also widely used, and it forms a disulfide linkage with cysteine. This linkage can be cleaved under reducing conditions, whereas PEG-Mal gives stable conjugates. PEG-IA and PEG-VS are both less reactive and infrequently used, whereas PEG-Mal and PEG-OPSS yield quantitative protein modification. Very recently, an interesting strategy was devised to direct PEGylation to protein disulfide bridges as well; in this case, the disulfide link is firstly reduced and then the resulting free thiols are reacted with a special PEG monosulfone reagent to give a stable three-carbon PEGylated bridge. This procedure, although very promising, is not described here further because the relevant PEG monosulfone is currently unavailable commercially. The reader, however, may refer to the literature for further details (23). 1.2.4. PEGylation to Carboxylic Acid Groups

The direct coupling of PEG-NH2 to activated protein carboxylic groups cannot be easily performed because it typically yields intra- or intermolecular linkages between the protein amines. An original solution uses PEG-hydrazide that is reactive at low pH toward carboxyl groups, but does not react with protein amino groups that are protonated under such conditions (24).

1.2.5. PEGylation of Proteins Modified with Aldehydic and Keto Groups

Aldehydic and keto groups, absent in natural proteins, can be exploited for certain types of nucleophilic additions. Aldehyde functional groups can usually be introduced into proteins by the oxidation of an N-terminal threonine or serine residue using sodium periodate. The introduced aldehydic groups can react with an aminooxy-functionalized PEG chain to obtain a conjugate at the N terminus of the protein (Fig. 7). In some cases, when the N terminus is an amino acid other than threonine or serine, a similarly reactive group can be introduced by metal-catalyzed oxidation, although the conditions for this reaction are potentially more damaging to proteins (25).

1.2.6. Selective PEGylation in Structuring or Denaturing Media

Although proteins typically present a defined structural conformation in aqueous solutions, peptides often have a random coil structure. The presence of organic solvents or chaotropic salts can promote structural rearrangements of the peptide and eventually modify the degree of solvent exposure and reactivity of some residues. As a few examples of selective PEGylation in the presence of organic solvents have been reported in the literature, they cannot be considered to be of general applicability.

Fig. 7. Conjugation of PEG to proteins containing aldehydic groups.

Covalent Conjugation of Poly(Ethylene Glycol) to Proteins and Peptides

103

In fact, several parameters such as the solvent, PEG/peptide molar ratio, peptide concentration, and the temperature need to be evaluated on a case-by-case basis. For instance, it was reported that in the presence of 60% (v/v) of dimethylformamide (DMF) at high pH, insulin is selectively modified by a NHS-activated PEG at LysB29 only, even though there are three potential sites for conjugation that are present within the peptide sequence (viz., at GlyA1, PheB1, and LysB29) (26). Moreover, growth hormone releasing factor (GRF), bearing the amino acid residues Lys12 and Lys21 within the N-terminal region, has been conjugated to activated PEGs to yield equimolar amounts of two distinct monoPEGylated isomers (i.e., Lys12- and Lys21-conjugates) (27). However, when the same reaction is conducted in trifluoroethanol (TFE) (50% v/v), it was found that 90% of the reaction mixture is composed of the derivative at Lys21. And finally, the lone Cys17 of G-CSF – buried within a hydrophobic region – could be PEGylated only under mild reversible denaturation conditions that preserved the integrity of the disulfide bridges (28). 1.2.7. Amino PEGylation of Peptides by Reversible Protection

For the selective PEGylation of peptides, a different type of strategy involves the reversible protection of specific residues. This procedure is possible for peptides only because they generally contain just a few nucleophilic groups and are more stable than full-length proteins toward harsh chemical treatments. This method involves three steps: (1) protection by suitable reagents of the residues known to be important for the activity and, eventually, the purification of the desired isomers; (2) PEGylation at the level of the lone unprotected, reactive target residue; and (3) removal of all the protecting groups. This method is not suitable for proteins because the harsh methods employed during the protection and/ or deprotection reactions can negatively affect protein integrity. The protected peptides can be obtained by solid-phase synthesis or by chemical modification. For example, a somatostatin analogue, bearing one terminal a-amino group and one lysine residue, was modified by selective pH-driven tert-butyloxycarbonyl (Boc) protection of the first amino group, followed by PEGylation of the second one and deprotection (29). In addition, GRF and salmon calcitonin, prepared by solid-phase synthesis, were fluorenylmethyloxycarbonyl (Fmoc)-protected at the N terminus and at one of the two internal lysine residues, and then selectively conjugated to PEG (30, 31).

1.2.8. Glutamine Enzymatic PEGylation

As an alternative to chemical conjugation, promising selective methods have been proposed that use enzymes to catalyze the covalent attachment of polymers to proteins. Among these, transglutaminase (TGase)-mediated conjugation has drawn significant attention for its high degree of site-specificity. TGase catalyzes an acyl transfer between the g-carboxamide group of a glutaminyl residue

104

Mero et al.

Fig. 8. PEGylation mediated by transglutaminase (TGase).

(acyl donor) and a primary amine (acyl acceptor). The latter can be selected among a variety of amines, including the e-amino group of lysine or an appropriate PEG derivative bearing an amino group (PEG-NH2) (Fig. 8) (32). Among the various prokaryotic and eukaryotic TGases that have been explored for conjugation applications, the most widely used enzyme is microbial TGase, which has a number of advantageous properties over eukaryotic TGases such as a calcium-independence and lower substrate specificity requirements. These properties conveniently allow for the use of TGase as a biochemical reagent on a large scale for industrial applications. In the area of pharmaceutical biotechnology, several proteins have been selectively modified by TGase; these are recombinant human interleukin-2 (IL-2), G-CSF, and human growth hormone. In these examples, the selective conjugation is due to the TGase active site structure, which is accessible only to those glutamines present within flexible regions of the protein substrate (33). 1.3. PEG-Conjugate Purification

Usually PEG coupling reactions, especially random amino PEGylation, yield a mixture of heterogeneous compounds and, therefore, a purification step is always required. Even in the case of a selective PEGylation reaction, a purification step is still needed to eliminate unreacted proteins, excess amounts of polymer, and various by-products. Several purification strategies may be used depending upon the properties of both the protein and the conjugated PEG moiety. Dialysis or ultrafiltration of the reaction mixture can be used to remove low-molecular weight components or to exchange the solvent. The removal of unreacted PEG polymers, non-conjugated proteins, and the separation of different PEGylated products can be achieved by using more specific and elaborate chromatographic techniques equipped with an online detector. Proteins and their conjugates can usually be monitored by measuring the UV absorbance at l = 214 or 280 nm or by fluorescence detection (lEX = 295 nm, lEM = 310 nm). On the contrary, unreacted PEGs, which are nearly transparent in the UV spectrum, can be monitored by using refractive index (RI) detectors (34) or light scattering techniques. Typically, the amount of PEG polymers present in a sample can be estimated by analyzing the collected fractions with an iodine assay (35).

Covalent Conjugation of Poly(Ethylene Glycol) to Proteins and Peptides

105

In Subheadings 3.3 and 3.4, we describe several protocols that can serve as a useful guideline for the purification and analysis of most types of PEGylated proteins. 1.3.1. Ion Exchange Chromatography

When PEG is coupled to a protein’s amino groups, the resulting conjugate may have an isoelectric point (pI) different from that of the starting native protein. Cation exchange chromatography (CEX) is the method of choice for the separation of PEGylated proteins because it exploits differences in charge at the protein surface (36, 37). Effectively, PEG modifies the elution time of proteins in ion exchange chromatography (IEX) separations, either by coupling to the amines or shielding the charges on the protein surface. Even in those situations where the net charge is unaltered with respect to the starting protein, the presence of the PEG moiety may decrease the interaction between the protein and the chromatographic matrix, thus yielding a shorter elution time for the PEG–protein conjugate compared to the starting protein. In the case of CEX separations, the elution order of PEG–protein conjugates is determined mainly by the number of linked PEG chains: Highly PEGylated molecules tend to elute first, followed by less PEGylated isomers and then by the unreacted protein (38); and any unreacted PEG that does not present positive charges elutes in the column void volume. Given that PEG–proteins tend to interact weakly with IEX matrices, special consideration must be given to the buffer conductivity and pH conditions. For this reason, it is convenient to carry out an extensive dialysis of the PEGylation mixture against the buffer that will be used for the CEX step in order to reach the same ionic strength (as measured by conductometry) and pH value of the column equilibration buffer. In addition, the reaction mixture samples should be diluted and filtered to avoid high back pressure and column fouling. Occasionally, a double-step procedure may be suitable, where a first chromatographic step is performed with a large-particle size resin to eliminate the free polymer, followed by a second step in which a resin with higher resolution is employed. The choice of buffers is a very important step for successfully carrying out IEX separations. Usually the equilibration buffer (A) has a low ionic strength, whereas the elution buffer (B) contains a higher concentration of salt (NaCl). In some cases, a change in pH can be also used between buffers A and B. Small increases in the salt concentration or pH of the buffer solution can effectively reduce the strength of the interactions between PEG–proteins and the resin, causing the conjugates to be eluted from the column before the un-PEGylated proteins. A problem that is commonly encountered in the purification of PEGylated proteins is the low capacity of the pores of conventional IEX media; the effective diameter of these pores can sometimes be too small to allow the penetration of

106

Mero et al.

PEG–protein complexes with a large surface area. Consequently, the desired conjugate products can be lost in the column flow-through. This is not generally considered to be an issue in analytical-scale experiments where the sample loading is relatively low (around 1  mg of protein/mL of sample); for preparative-scale purifications, however, a higher loading (e.g., >6 mg/mL) of the PEG– protein mixture may become a limiting factor. Conventional strong cation exchange resins (e.g., Mono S™ (sulfo) or Mono SP™ (sulfopropyl) sepharose (GE Healthcare)) that are usually employed with standard fast protein liquid chromatography (FPLC) or HPLC systems can be loaded at a protein concentration of 5–10  mg/mL at a linear flow rate of 60  cm/h. MacroCap™ SP (GE Healthcare) is a new type of strong cation exchange medium especially designed for the purification of PEGylated proteins and other large biomolecules under high sample loading conditions (12–15  mg/mL). The matrix is highly porous and provides good mass transfer characteristics and improved accessibility to the internal surface area for the adsorption of large molecules. PEGylated proteins can be also purified by anionic exchange chromatography (AEX) (40). In this case, the PEG–protein conjugate is bound to the column at a higher pH value than its isoelectric point (pI), which results in a negative net charge on the molecule. Once again, the charges at the surface of the protein can be shielded by the presence of PEG. As in CEX, it is often convenient to use a linear salt or pH gradient to elute the PEGylated derivatives from the column. 1.3.2. Reversed-Phase Liquid Chromatography

Reversed-phase high performance chromatography (RP-HPLC) is often used for the characterization of PEGylated species due to its high resolution and the possibility of coupling the technique with an online mass spectrometer detector (41). Although the method is very rapid, only a limited amount of sample material can be loaded at once. Furthermore, PEGylated proteins often give wide peaks due to polymer polydispersity, which compromises the resolution of PEGylated species to different extents. The elution conditions employed in RP-HPLC generally require high percentages of acetonitrile (CH3CN) or methanol (MeOH), which can be a concern for protein stability. Other parameters such as the column temperature, the elution gradient profile, and the mobile phase composition must all be carefully evaluated on a case-by-case approach.

1.3.3. Size-Exclusion Chromatography

Size-exclusion chromatography (SEC) separates molecules based on differences in their hydrodynamic volumes. Since PEGylation is often performed with the aim of increasing the size of proteins, it is clear that SEC can be a useful technique for purifying PEG– protein conjugates. SEC, however, has several limitations: it generally gives broad peaks with poor resolution for PEG-conjugates; it

Covalent Conjugation of Poly(Ethylene Glycol) to Proteins and Peptides

107

is a low-throughput technique with relatively high costs, and thus has limited applicability in large-scale processes; and it also cannot separate positional isomers (which have the same mass and hydrodynamic volume). Furthermore, in those cases where there are only small differences in size, unreacted PEG and protein molecules can be co-eluted with monoPEGylated species. Conversely, SEC is a useful method for removing low-molecular weight impurities (e.g., by-products formed by the hydrolysis of functionalized PEG, buffer salts, solvents, and other low-molecular mass reagents), and also serves as an effective tool for obtaining an initial evaluation of the degree of protein modification as well as for determining the presence of aggregates. Typically, dextran- and agarose-based SEC columns are often used in conjunction with FPLC systems, with Sephadex G-75, G-50, and Superose 6 or 12 (42) being the most commonly used media to purify proteins. The choice of a particular column type depends on the molecular weight of the protein. Generally, buffers with a low percentage of organic solvents are used to minimize the occurrence of hydrophobic interactions between proteins in the sample mixture and the column matrix (43–45). One important factor in the use of SEC for PEGylation applications is to take into account that the apparent size of a PEG–protein conjugate is roughly five to ten times larger than that of a globular protein with the same nominal molecular weight. 1.3.4. Ultrafiltration/ Diafiltration

One particularly important nonchromatographic step in the production of pharmaceutical proteins is the ultrafiltration/diafiltration operation. Ultrafiltration/diafiltration are effective processes for exchanging buffers between chromatographic steps and for concentrating the conjugate products to achieve the desired final concentration. Usually, membrane filters with the same nominal molecular weight cutoff value as those used for the native (unmodified) protein should also be employed to concentrate monoPEGylated-protein samples since such types of PEGylated products can potentially escape by the so-called “snake effect” through membranes with larger porosities, causing substantial losses of product (46–48). In the case of branched PEG or multiPEGylated conjugates, larger cutoffs may be used, since in this case the “snake effect” of PEGs is not as likely to be an issue. Regenerated cellulose and polyethersulfone membranes, with their low protein binding and high protein retention characteristics, are the most commonly used filter media and also provide the highest rates of recovery for purified proteins (48). However, polyethersulfone filters are much less efficient for processing PEGylated protein species because of the hydrophobic nature of their membrane surfaces. Indeed, the full application potential of membrane-based technologies for PEGylated products is somewhat limited because the increased size, greater hydrophobicity,

108

Mero et al.

and lower electrostatic interactions of PEG-conjugates with respect to unmodified proteins tend to lead to increased fouling, i.e., the largely irreversible adsorption and/or deposition of proteins on and within membrane filter media (49). 1.4. PEG-Conjugate Characterization

The first issue to be faced with the characterization of PEGconjugates deals with the accurate determination of the amount of attached PEG. Several methods are available such as colorimetric assays, SEC, electrophoresis, and mass spectrometry (MS). However, these methods are not suitable for separating and identifying positional isomers of PEG; for these purposes, IEX is generally better suited. The identification of PEGylation sites within proteins is based mainly on the use of standard protein sequence analysis methods, with peptide mapping and mass spectrometry both falling within the scope of such types of useful techniques. The simplest and most rapid methods available for conducting preliminary characterizations of a PEG–protein conjugate are colorimetric assays. In the case of amino-PEGylation products, the Habeeb assay can be used to quantify the amount of unreacted protein primary amines with TNBS reagent, thus allowing one to indirectly calculate the number of bound PEG chains (50). In a similar approach, Ellman’s assay can be used to determine the presence of any remaining free cysteines following PEG–thiol conjugation (51). Finally, the iodine assay, based on the noncovalent interaction of iodine with the PEG backbone, can be used to obtain both qualitative and quantitative information about the polymer (35). The preceding methods are all described in detail in the following sections. To characterize the protein conformation in the conjugates, various spectroscopic methods may be used. In particular, circular dichroism, fluorescence, and UV spectra can all be applied to analyze PEG–protein conjugates – thanks to the fact that PEG is transparent to light at UV–visible wavelengths. For the interpretation of such spectra, the reader is referred to dedicated books on protein characterization (52).

1.4.1. Bicinchoninic Acid Assay

The protein concentration in a non-PEGylated sample can usually be determined by simply measuring the spectrophotometric absorbance of the aromatic amino acids residues at l = 280 nm. This approach may also be acceptable for measuring the concentration of a PEG–protein conjugate sample after first confirming that the absorbance of the protein is not altered by PEG coupling. Alternatively, a colorimetric assay can be carried out to estimate the protein concentration. The bicinchoninic acid (BCA) assay, for example, is the most commonly used technique because it is less affected than other types of dye-binding assays by the presence of PEG. BCA protein assay kits are commercially available from a number of different suppliers (e.g., Sigma-Aldrich, Bio-Rad,

Covalent Conjugation of Poly(Ethylene Glycol) to Proteins and Peptides

109

Thermo Fisher, etc.); in these assays, the quantitative determination of the protein concentration in a PEGylated sample relies on the generation of a calibration curve using standard solutions of the native protein (53). 1.4.2. Ion Exchange Chromatography

IEX, as previously discussed in Subheading  1.3.1, is the most widely used technique for the fractionation and purification of PEGylated proteins on a preparative scale. IEX is also very useful for analytical purposes because it allows the efficient separation of positional isomers (54). In particular, our laboratory has found that analytical strong cation exchange columns (e.g., TSKgel SP-5PW, 7.5  mm × 7.5  cm (Tosoh); and HiLoad™ 16/10 SP Sepharose™ HP (GE Healthcare)) can provide good results for PEGylated samples.

1.4.3. Reversed-Phase High Performance Liquid Chromatography

RP-HPLC is a useful technique for the determination of purity and species content of PEGylated protein samples. This fractionation method is based on differences in hydrophobicity of the native and PEGylated proteins. PEG is an amphiphilic molecule, and thus PEGylated proteins often exhibit higher retention times on RP-HPLC columns than their un-PEGylated counterparts. The increase is generally dependent on the length and mass of the conjugated polymer (37). RP-HPLC is not only an excellent and robust tool for the fractionation of PEGylated and un-PEGylated species, but is also a useful method to detect protein oxidation, deamidation, or cleavage of the protein backbone (55). In addition, the technique can also be used for the high-resolution analysis of protein fragments and peptides (i.e., peptide fingerprinting) (19, 43), or to separate positional isomers (53) in the case of peptide PEGylation. In analytical RP-HPLC applications, the PEG/protein ratio appears to be the predominant factor affecting the resolution of PEGylated conjugates (37). Typically, reversed-phase columns containing packings such as butyl (C4) or octadecyl (C18) can be employed for the fractionation of PEGylated protein using an elution gradient of H2O/CH3CN (either with or without trifluoroacetic acid (TFA)).

1.4.4. Size-Exclusion Chromatography

SEC, or gel filtration, is widely used to estimate the molecular weight (MW) of native (unmodified) proteins through the use of a standard calibration curve. As PEGylated proteins show a larger hydrodynamic volume than native proteins of the same nominal MW, SEC (and, similarly, sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE)) cannot provide an accurate determination of the exact MW of PEG–protein conjugates, but can only be used to monitor a PEGylation reaction and characterize the homogeneity of the conjugate product. An interesting and useful discussion on the effect of PEG on the apparent size of conjugates and their behavior in SEC has recently appeared in the literature (57).

110

Mero et al.

Several brands of gel filtration columns are commercially available in the marketplace such as Superose 6 or 12 (GE Healthcare), TSK-Gel (Tosoh), Zorbax GF-250 (Agilent), and BioSep SEC (Phenomenex). Each type of column has a specified MW fractionation range, and hence the selection of the most appropriate column to use must be done on a case-by-case. Typically, phosphate and HEPES buffers containing a salt (NaCl) gradient are the most often used eluents in SEC applications. The addition of water-miscible organic solvents (e.g., acetonitrile or isopropanol) into the mobile phase can generally improve the overall separation of PEG–protein conjugate mixtures by increasing peak sharpness and reducing peak tailing due to nonspecific adhesion of PEGs to the stationary phase (43–45). SEC-HPLC is also useful for the determination of free PEG in PEGylated samples, but since free PEG itself does not absorb at UV wavelengths, it is necessary to use a refractive index (RI) detector coupled with a second UV detector to reveal simultaneously the protein fractions, PEG, and other low molecular weight by-products (e.g., NHS) that emerge from the column. It should be noted, however, that a RI detector cannot be used for analyses requiring a gradient elution because the changes in the eluent composition modifies the baseline. 1.4.5. Sodium Dodecyl Sulfate Polyacrylamide Gel Electrophoresis

SDS-PAGE is a highly useful technique for determining the purity and molecular weight (MW) of native (unmodified) protein samples. However, SDS-PAGE is not suitable as a direct method to easily evaluate the MW of PEGylated proteins because proper calibration standards are typically unavailable for such analyses. During electrophoresis, the migration rate of PEG-conjugates through the porous gel matrix is significantly slowed by the long and heavily hydrated PEG chains. Consequently, PEGylated proteins usually display apparent MW values on SDS-gels that do not correlate with that of the free protein MW standards (54). Generally, SDS-PAGE can only be used to qualitatively follow the progress of a PEGylation reaction and the subsequent purification procedures. For example, a significant shift in the SDS-PAGE band position for a protein after PEGylation would provide evidence in support of a success PEGylation experiment. Random multi-PEGylation will typically yield several bands on an SDSPAGE gel, with each band migrating slower than that corresponding to the native protein; on the contrary, only a single new band should be expected to appear for monoPEGylated derivatives. In some cases, using PEG molecules of different MWs instead of native protein standards may enable a rough estimation of the apparent MW of the conjugates (58, 59); however, it is then necessary to use a specific iodine staining procedure to reveal the PEG bands since PEG is negative toward conventional protein stains. SDS-PAGE gels of varying degrees of crosslinking

Covalent Conjugation of Poly(Ethylene Glycol) to Proteins and Peptides

111

can usually be employed for such analyses, but attention must be paid to the full range of MWs actually present in a particular PEGylated sample in order to accommodate the migration of all the components in the mixture. 1.4.6. Mass Spectrometry of Conjugates

To evaluate the molecular weight of PEGylated proteins, mass spectrometry (MS) analyses are highly recommended. Typically, MALDI-MS is employed for such determinations, even if these types of spectra are sometimes complicated by PEG polydispersity. For PEGylated proteins, the total MW is easily calculated as the sum of the MW of the native protein and the MW of the conjugated PEG chains (60). Nevertheless, the MS technique does present a few limits, such as (1) poor ionization efficiency with large polymers, which can affect the detection sensitivity as well as the accurate determination of the average MW and (2) degradation of the proteins during sample extraction and evaporation.

1.4.7. General Method for Determination of PEGylation Sites for Proteins Modified with Polydisperse PEGs

The method used for the localization of the sites of PEG conjugation follows the strategy commonly employed for conventional protein sequence determinations. This approach is based on the proteolytic digestion of the native (unmodified) protein and the conjugated protein, followed by a comparison of the two elution patterns obtained by RP-HPLC. Subsequent analysis of each peak in the RP-HPLC fingerprint by ESI-MS or MALDI-MS allows the determination of peptide composition (61, 62) and the site of amino acid PEGylation. The determination is carried out by comparison of the fingerprint of the native protein and that of the conjugate. The peptides that are missing in the conjugate elution pattern represent those sequences that contain the polymer. Typically, trypsin is the most commonly employed proteolytic enzyme; however, other proteolytic enzymes with different digestion specificities may also be used depending on the sequence of the protein under investigation. This is necessary in those cases where the PEGylated peptide fragments contain more than one potential site of PEGylation.

1.4.8. Simplified Method for Determination of PEGylation Sites for Proteins Modified with Monodisperse PEGs

The method described above in Subheading 1.4.8 is an “indirect” analysis because, due to the polydispersity of the polymer, the PEGylated peptide is identified by the disappearance of the corresponding fragment. A simplified procedure for the identification of PEGylation sites can be applied if a monodisperse PEG is used for conjugation. Monodisperse PEG polymers, only recently available in the marketplace, are detectable by ESI-MS, as well. Unfortunately, monodisperse PEGs are available only in low MW forms, making them unsuitable for the half-life prolongation of proteins; however, they can be useful for characterization purposes. A monodisperse PEG was recently employed by our group for the specific modification of several proteins of pharmaceutical

112

Mero et al.

interest mediated by the enzyme TGase (63). ESI-MS and tandem mass spectrometry (MS/MS) of the digested peptides allowed for a “direct” (as opposed to indirect, as above) characterization of the peptide and the identification of the PEGylated site even if more than one is present within the same peptide. It is important to note that both the high-MW polydisperse PEG and the low-MW monodisperse form of the polymer are linked to the same site in the protein sequence since the specificity of the conjugation site is dictated by the TGase enzyme (31, 61). Generally, the analysis procedure requires only a very small amount of sample material due to the possibility of determining both composition and sequence in a single analysis.

2. Materials 2.1. Characterization of PEGylating Agents

1. PEG reagents (see Note 1). 2. Deuterated DMSO-d6, CDCl3, or water (D2O). 3. Borate buffer, pH 8: 0.2 M borate buffer, pH 8. 4. Borate buffer, pH 9.3: 0.1 M borate buffer, pH 9.3. 5. Gly-Gly solution (2 mM): Dissolve 28.5 mg of glycyl-glycine in 100 mL of 0.2 M borate buffer, pH 8. 6. TNBS (1% w/v solution in DMF). 7. Phosphate-ethylenediaminetetraacetic acid (EDTA) buffer: 0.1 M Sodium phosphate buffer, 1 mM EDTA, pH 7. 8. Cys or GSH solutions (2 mM): Dissolve 2.61 mg of Cys or 6.62 mg of GSH in 10 mL of phosphate-EDTA buffer, pH 7. 9. Ellman’s reagent (10 mM): Dissolve 4 mg of DTNB in 1 mL of 0.1 M phosphate-EDTA buffer, pH 7. 10. MALDI-TOF MS matrix: Sinapinic acid (58), dihydroxybenzoic acid (65), or a-cyano-4-hydroxycinnamic acid (46) mixed with acetonitrile/water (1:1, v/v).

2.2. Protein PEGylation

1. Glycine solution: Use 200 mL of glycine (250 mM) in water, pH 7.4 for each milliliter of the reaction mixture. 2. NaCNBH3 solution: Prepare 20 mM NaCNBH3 in the same buffer as that used in the reaction mixture. 3. Hydroxylamine solution: Prepare 2 mL NH2OH, pH 7.3 in water.

2.3. Conjugate Purification

1. CEX-buffer A: 10 mM phosphate buffer, 10 mM NaCl, pH 4.7. 2. CEX-buffer B: 10 mM phosphate buffer, 0.1 mM NaCl, pH 4.7. 3. RP-buffer A: 0.05% (v/v) TFA in water. 4. RP-buffer B: 0.05% (v/v) TFA in acetonitrile (CH3CN).

Covalent Conjugation of Poly(Ethylene Glycol) to Proteins and Peptides

2.4. Conjugate Characterization

113

1. BCA Protein Assay Kit (Pierce): To prepare the BCA working solution, mix 50 parts of reagent A (containing sodium carbonate, sodium bicarbonate, BCA, and sodium tartrate in 0.1 M sodium hydroxide) with one part of reagent B (containing 4% CuSO4·5H2O) according to the manufacturer’s instructions. 2. BaCl2 solution: 5% (w/v) barium chloride in 1N HCl. 3. Iodine solution: Dissolve 1.27 g of I2 in 100 mL of 2% (w/v) KI in water. 4. Phosphate buffer: 20 mM phosphate buffer, pH 7.2. 5. Bicarbonate buffer: 4% (w/v) NaHCO3 buffer, pH 8.5. 6. Tris-Gdn buffer: 6 M guanidine hydrochloride, 50 mM Tris– HCl, pH 9.0. 7. PepClean™ C-18 spin columns (Pierce, Rockford, IL).

3. Methods 3.1. Characterization of PEGylating Agents 3.1.1. NMR Spectroscopy of PEGylating Agents

3.1.2. Analysis of PEG Diol Content in mPEG Batches

1. Dissolve the PEG sample (10–20 mg) (see Note 1) in 0.75 mL of deuterated solvent and perform the NMR analysis according to the instrument manufacturer’s instructions. In the 1H NMR spectrum, the integral values of the reactive group signals can be compared with the integral values of the backbone chain signals (–CH2–CH2–; e.g., PEG 5 kDa, 3.6 ppm, 491H) or with other characteristic signals of the polymer. For example, Fig.  1 shows the NMR spectrum of commercial BocPEG-NHS (5  kDa), where the H signals of the Boc group (–C(CH3)3, 1.4 ppm, 9H) are compared with those arising from the N-hydroxysuccinimide group (–NHS) (–CH2–CH2, 1.2 ppm, 4H) and the backbone chain signals. In this case, the analysis of the integrals indicates that the polymer is activated with NHS at only 60% of the maximum level, since the signal integration value is 2.4 instead of the expected value of 4. 1. Equip an HPLC system with a suitable SEC column (see Subheading  1.4.4 for a discussion on column selection). Equilibrate the column with the desired elution buffer. 2. Solubilize the PEG sample (0.2  mM) in 1  mL of elution buffer. 3. Load 20 mL of the sample solution onto the SEC column. 4. PEG does not absorb at wavelengths suitable for UV–visible detection and can be revealed by employing a refractive index detector, instead.

114

Mero et al.

Table 2 Preparation of test solutions for the TNBS assay Blank

PEG reaction mixture

Gly-Gly standard solution

20 mL of TNBS

20 mL of TNBS

20 mL of TNBS

980 mL of Borate pH 9.3

955 mL of borate pH 9.3

955 mL of borate pH 9.3

25 mL of Sample A

25 mL of Sample G

3.1.3. Evaluation of the Degree of Activation of Amino-Reactive PEGylating Agents

1. Prepare Sample A (PEG reaction mixture) as follows: To 1 mL of 2 mM Gly-Gly solution, add 1 eq. (2 mmol) of the activated PEG polymer (e.g., PEG-NHS). The required amount of activated PEG to add depends on its MW; for example, if the MW of PEG-NHS is 5 kDa, then 10.8 mg of the activated polymer should be added to 1 mL of 2 mM Gly-Gly solution. 2. Let the mixture react for 30 min at room temperature under continuous agitation. 3. Prepare 1 mL of 2 mM Gly-gly solution (sample G) as control. 4. Perform the TNBS assay in duplicate at room temperature according to Table 2. 5. Incubate the reactions for 30 min. 6. Read the absorbance at l = 420  nm using a UV–visible spectrophotometer.



The percentage of activation of the amino-reactive PEGs is calculated by using the following formula: % Activation = [1 - (AbsA - AbsB) / (AbsG - AbsB)] ´ 100% AbsA = Absorbance of PEG reaction mixture AbsG = Absorbance of Gly-Gly standard solution AbsB = Absorbance of the blank solution.

3.1.4. Evaluation of the Degree of Activation of Thiol-Reactive PEGylating Agents

1. Prepare Sample A (PEG reaction mixture) as follows: To 1 mL of 2 mM Cys or GSH solution, add 1 eq. of the thiolreactive PEG molecule. The required amount of thiolreactive PEG to add depends on its MW; for example, if the MW of the PEG is 5 kDa, then 10.2 mg of polymer should be added to 1 mL of 2 mM Cys or GSH solution. 2. Let the mixture to react for 30  min at room temperature under continuous agitation. 3. Prepare 1 mL of 2 mM Cys or GSH (Sample C) as control. 4. Perform Ellman’s assay in duplicate at room temperature according to Table 3. 5. Incubate the reactions for 15 min. 6. Read the absorbance at l = 412  nm using a UV–visible spectrophotometer.

Covalent Conjugation of Poly(Ethylene Glycol) to Proteins and Peptides

115

Table 3 Preparation of test solutions for Ellman’s assay Blank

PEG reaction mixture

Cysteine standard solution

50 mL of Ellman’s reagent

50 mL of Ellman’s reagent

50 mL of Ellman’s reagent

1 mL of phosphateEDTA pH 7

970 mL of phosphateEDTA pH 7

970 mL of phosphateEDTA pH 7

30 mL of Sample A

30 mL of Cys or GSH solution

The percentage of activation of thiol-reactive PEGs is calculated by using the following formula:

% Activation = [1 - (AbsA - AbsB) / (AbsC - AbsB)] ´ 100% AbsA = Absorbance of PEG reaction mixture AbsC = Absorbance of cysteine standard solution AbsB = Absorbance of the blank solution.

3.1.5. Half-Life Measurement of NHSActivated PEGs

1. Prepare a solution of NHS-activated PEG (0.2–0.5  mM) in dioxane. 2. Add 50 mL of the PEG-NHS solution to 950 mL of 0.2 M borate buffer (pH 8.0) and immediately read the absorbance at 280 nm every 5 s until a plateau is reached. To evaluate the aminolysis rate, add 50 mL of the PEG-NHS polymer solution to 950 mL of a Gly-Gly solution (Gly-Gly/PEG-NHS polymer, 1:1 molar ratio) in 0.2 M borate buffer (pH 8.0). 3. The absorbance (l = 280 nm) at time point = 0 s is taken to be the blank and corresponds to the NHS already present in the mixture. 4. Plot the resulting absorbance data and determine the half-life of the PEG-NHS reagent.

3.1.6. Mass Spectrometric Analyses of PEGylating Agents

1. Dissolve PEG (20 mg) in 0.1% (v/v) TFA in water (5–10 mL). 2. Mix one volume of saturated matrix solution with one volume of the PEG solution. 3. Load the matrix/PEG mixture (10–20 mL) onto the MALDI sample plate and perform the MS analysis according to the instrument manufacturer’s instructions.

3.1.7. Analysis of Low-Molecular Weight Impurities of PEG-Aldehyde

1. Dissolve PEG-aldehyde (0.25–1 mM) in water. 2. At determined time points, withdraw 100  mL and treat the sample aliquot with 2,4-dinitrophenylhydrazine (DNPH/ PEG-aldehyde, 1:1 eq.).

116

Mero et al.

3. Fractionate the resulting reaction mixture by RP-HPLC on a C8 column (100 × 2.1 mm; flow rate = 0.8 mL/min; mobile phase = H2O/CH3CN/H3PO4, 1:1:1). 4. Collect the fractions, lyophilize, and analyze by RP-HPLC on a C18 column (flow rate = 0.2 mL/min; mobile phase = H2O/ CH3CN) using an ESI-MS detector operating in the negative ion mode to identify low-MW impurities. 5. As an alternative to the above procedure (Steps 1–4), dissolve the PEG-aldehyde in MeOH and infuse the sample into an ion mobility/quadrupole/time-of-flight mass spectrometer at a flow rate of 5 mL/min to analyze for the presence of truncated PEG-aldehyde molecules. 1. Prepare the protein solution (1–5 mg/mL) in 0.1 M borate buffer, pH 8.0–9.0.

3.2. Protein PEGylation 3.2.1. Random PEGylation at Free Amino Groups of a Protein

2. Determine the exact concentration of the protein by UV absorption using its molar extinction coefficient (66). 3. Add the amino-reactive PEG – in small amounts – to the protein solution under gentle stirring. An excess of activated PEG is usually required. The optimum ratio of PEG to each protein amino group may range from 1 to 10, depending on the particular PEG and the reactivity of the amino groups on the protein. Table 4 lists some examples of protein conjugation experiments that have reported in the literature using different ratios of PEG/protein, PEGs with different MWs and different protein concentrations. 4. Incubate the reaction mixture at room temperature for 1–5 h. 5. Quench the reaction with glycine solution and stir for 1 h. 6. Eventually add 1 mL of hydroxylamine solution and stir for 30 min.

Table 4 PEGylation reaction conditions and yields as reported in the literature for different molar ratios of NHS-activated PEG to protein NH2 groups Reaction conditions

Protein concentration Protein MW PEG MW PEG/protein (mg/mL) (kDa) (kDa) molar ratio Yield (%) References

Aqueous, pH 8.5

10

21

5

1–3

85

(42)

Aqueous, pH 8.5

4

23

5

5

50

(67)

Aqueous, pH 8.5

1.5

13.7

5

2.5

55

(67)

Aqueous, pH 8.5

6

10

3

50

(68)

Aqueous, pH 8.5

2

17.3

5

8

60

(42)

Aqueous, pH 9

5

19.3

40

3

60

(69)

141

Covalent Conjugation of Poly(Ethylene Glycol) to Proteins and Peptides

117

7. Dialyze the resulting solution to eliminate low-molecular weight products. 8. Purify the PEG–protein conjugates as described in Subheading 3.3, and characterize the products as described in Subheading 3.4. 3.2.2. PEGylation at the N Terminus of a Protein Using PEG-Aldehyde

1. Prepare a protein solution at 1–5 mg/mL (as determined by absorbance spectrophotometry) in a buffer with pH ~5.0–6.0. 2. Add PEG-aldehyde (see Note 2) to the protein solution at the desired molar ratio. Depending on the protein properties, a great excess of PEG-aldehyde may be needed. It is advisable to test several different PEG/protein molar ratios to optimize the reaction (e.g., 10–50 eq. with respect to the amount of protein molecules). 3. Incubate the reaction mixture for 1 h, and then add NaCNBH3 solution (20 mM) (add 50 eq. of NaCNBH3 per 1 eq. of PEG). 4. Incubate the reaction mixture further at 4°C for 24 h under gentle stirring. 5. Quench the reaction with glycine solution and stir for 1 h. 6. Purify the PEG–protein conjugates as described in Subheading 3.3, and characterize the products as described in Subheading 3.4.

3.2.3. Thiol PEGylation

1. Prepare a protein solution of 1–5 mg/mL in 0.1 M phosphateEDTA buffer, pH 7.2. 2. Add PEG-OPSS (see Note 3) or PEG-maleimide PEG-MAL (see Note 4) to the protein solution at a molar ratio 1:1 or 2:1 with respect to the amount of free thiols present. If PEG-VS is used, an excess of 2–10 eq. is recommended. 3. Incubate the reaction mixture at 4°C for 4–24 h (depending on the PEG derivatives used) under gentle stirring. 4. Monitor the disappearance of the free thiols in the reaction mixture by Ellman’s assay (see Subheading 3.4.7). 5. Purify the PEG–protein conjugates as described in Subheading 3.3, and characterize the products as described in Subheading 3.4.

3.2.4. PEGylation at Protein Carboxylic Groups

1. Prepare a protein solution at 1–5  mg/mL in 0.1  M phosphate buffer, pH 4.0–5.0 (see Note 5); 2. Add 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (2–5 eq. with respect to the amount of protein carboxylic acid groups) and PEG-hydrazide (10–50 eq. with respect to the amount of protein molecules).

118

Mero et al.

3. Incubate the reaction mixture at 4°C for 24 h under gentle stirring. 4. Purify the PEG–protein conjugates as described in Subheading 3.3, and characterize the products as described in Subheading 3.4. 3.2.5. PEGylation of Proteins Containing Introduced Aldehydic Groups

1. Prepare a solution of protein containing N-terminal serine or threonine residues at 1–5 mg/mL in 1% (w/v) ammonium bicarbonate buffer, pH 8. 2. Add a tenfold molar excess of sodium periodate for 10 min. 3. Add a 2,000-fold molar excess of ethylene glycol to stop the oxidation reaction, and then dialyze the mixture against an acidic buffer solution. 4. Add aminooxy-PEG (10–50  eq.) and adjust the pH of the reaction mixture to 3.6. 5. Incubate the reaction mixture for 20 h at room temperature. 6. Purify the PEG–protein conjugates as described in Subheading 3.3, and characterize the products as described in Subheading 3.4.

3.2.6. Selective PEGylation of Peptides in Structuring or Denaturing Media

1. Dissolve the peptide (2 mg/mL) in a suitable mixture of H2O and a water-miscible organic solvent (e.g., H2O/DMF, 2:3 v/v; or H2O/TFE, 1:1 v/v); 2. Bring the pH of the peptide solution to 9–10. 3. Add PEG-NHS (1–3 eq. with respect to the peptide) to the peptide solution. 4. Incubate the reaction mixture for 30 min under gentle stirring. 5. Quench the reaction mixture with glycine solution and continue stirring for 1 h. 6. Purify the PEG–protein conjugates as described in Subheading 3.3, and characterize the products as described in Subheading 3.4.

3.2.7. Amino PEGylation of Peptides by Reversible Protection

1. Protect the most reactive amine groups of the peptide with a suitable procedure (e.g., using t-Boc or Fmoc groups); 2. Purify the desired products by chromatography (e.g., RP-HPLC). 3. Dissolve the protected peptide in DMF, DMSO, or other solvent at a final concentration of 5–10 mg/mL. 4. Add an activated amino-reactive PEG at an excess of 2–10 eq. over the peptide. 5. Incubate the reaction mixture at room temperature for 4 h under gentle stirring. 6. Quench the reaction with glycine solution and continue stirring for 1 h.

Covalent Conjugation of Poly(Ethylene Glycol) to Proteins and Peptides

119

7. Dialyze the solution against water to remove the organic solvent and lyophilize. 8. Deprotect the conjugate using a suitable method; for example, Boc removal can be carried out by dissolution in TFA, while the Fmoc group can be cleaved in 20% (v/v) piperidine in DMF. 9. Purify the PEG–protein conjugates as described in Subheading 3.3, and characterize the products as described in Subheading 3.4. 3.2.8. Enzymatic PEGylation Using Microbial Transglutaminase

1. Prepare a protein solution at 1–5  mg/mL in 0.1  M phosphate buffer, pH 7.0. 2. Add an excess of 2–10 eq. of PEG-NH2 with respect to the amount of protein. 3. Add TGase at an enzyme:substrate ratio of 1:75 (w/w). 4. Incubate the reaction mixture at room temperature for 4 h under gentle stirring. 5. Quench the reaction with a few drops of acetic acid at pH 3. 6. Purify the PEG–protein conjugates as described in Subheading 3.3, and characterize the products as described in Subheading 3.4.

3.3. Conjugate Purification

3.3.1. Ultrafiltration/ Diafiltration

After quenching the PEGylation reaction mixture, dialysis is typically performed using regenerated cellulose membranes against buffers with low ionic strength at 4°C (see Note 6) for 2 days with continuous stirring. As an alternative to dialysis, ultrafiltration/diafiltration can also be used to exchange the buffer components of the reaction. For both procedures, it is always important to verify that the protein remains stable during all the processing steps. After dialysis or ultrafiltration/diafiltration, the removal of unreacted PEG, unreacted protein, and the separation of the different PEGylated species can be achieved by using several different chromatographic techniques, as described in the following sections. 1. Equilibrate the membrane filter in isopropanol for 45 min to remove any wetting/storage agents. Following this, wash the membrane with water and equilibrate it in the same buffer as that used in the PEG–protein solution mixture. 2. Add the PEG–protein solution and allow the system to equilibrate before pressurizing. 3. Fill the stirring cell with the feed solution and connect it to a reservoir containing pure buffer. 4. Pressurize the system with air and monitor the filtrate flux over time; use small adjustments of the pressure to maintain a constant flux. Samples can be removed periodically from both the collected filtrate and the stirred cell to analyze the solute concentration by SEC-HPLC in order to detect any loss of protein.

120

Mero et al.

3.3.2. Cationic Exchange Chromatography

1. Equilibrate a cation ion exchange column with CEX-buffer A (see Note 7) according to the manufacturer’s instructions. For a TSKgel SP-5PW column (21.5 mm × 15 cm, 10 mm), it is recommended to use a flow rate of 5–8 mL/min. 2. Load the dialyzed reaction mixture onto the column. 3. Elute the PEGylated products by slowly increasing the elution gradient with CEX-buffer B. 4. Analyze the eluted fractions for the presence of PEG by performing an iodine assay, and for the presence of protein by monitoring the UV absorption. Collect and pool the fractions containing the PEGylated products. 5. Concentrate and exchange the buffer against CEX-buffer A. Keep the solution at 4°C for short-term storage or at −20°C for long-term storage (see Note 8).

3.3.3. Reversed Phase HPLC

1. Connect C4 (or C18) RP-HPLC column to the HPLC system. 2. Equilibrate the C4 column (250 × 21.1  mm, 10  mm) with RP-buffer A at a flow rate of 8 mL/min. 3. A column temperature of ~45°C is recommended for running the RP-HPLC procedure. 4. Load the dialyzed reaction mixture onto the column at 5–10 mg/mL (total protein concentration). 5. Elute the column initially with RP-buffer A. 6. Continue the elution using a moderately shallow gradient (1–2% per min) with RP-buffer B (see Note 9). 7. Collect and pool the fractions containing protein (as detected by monitoring the UV absorbance). Check for the presence of PEG by performing an iodine assay. 8. Concentrate and exchange the buffer against an appropriate saline solution and keep at 4°C for short-term storage or −20°C for long-term storage (see Note 8).

3.3.4. Size-Exclusion Chromatography

1. Connect a suitable SEC column to the HPLC system. 2. Equilibrate the SEC column with an appropriate saline buffer for 1  h at a flow rate of 1  mL/min (for a 10 × 300  mm column). 3. Load the dialyzed reaction mixture at 5–10  mg/mL (total protein concentration). 4. Elute the product with the same buffer used for column equilibration. 5. Collect and pool the fractions containing protein (as detected by monitoring the UV absorbance). Check for the presence of PEG by performing an iodine assay.

Covalent Conjugation of Poly(Ethylene Glycol) to Proteins and Peptides

121

6. Concentrate the product by ultrafiltration and keep at 4°C for short-term storage or −20°C for long-term storage (see Note 8). 3.4. Conjugate Characterization 3.4.1. BCA Protein Assay

1. Prepare a series of unmodified (i.e., non-PEGylated) protein samples with known protein concentrations (0.2–1.2  mg/ mL) at a final volume of 50 mL. 2. Prepare triplicate samples (50 mL each) of the PEGylated protein with unknown concentrations. 3. To each sample, add 1 mL of BCA working reagent. 4. Incubate the samples for 30 min at 37°C and then read the absorbance at l = 562 nm. 5. Prepare a standard curve by plotting the measured absorbance values versus protein concentration. Using the standard curve, determine the protein concentration of the PEGylated samples.

3.4.2. Ion Exchange Chromatography

1. Pre-equilibrate the column (7.5 × 75 cm, 5-mm particle size) with CEX-buffer A at a flow rate of 1 mL/min. 2. Load 200  mL of the PEG–protein conjugates (1  mg/mL protein) dissolved in CEX-buffer A. 3. Elute the products with CEX-buffer B using a shallow elution gradient. 4. Analyze the eluate with a UV–visible or fluorescence detector at a suitable wavelength. 5. In case further analysis of the samples is desired (e.g., SDS electrophoresis, mass spectrometry), collect the fractions containing the conjugate and use dialysis or ultrafiltration to change the buffer. 6. Lyophilize the conjugate products and store at −20°C until use (see Note 8).

3.4.3. Reversed-Phase HPLC

1. Pre-equilibrate the column with RP-buffer A at a flow rate of 1 mL/min. 2. Load 20 mL of the PEG–protein conjugates (0.1 mg/mL of protein) solubilized or diluted in RP-buffer A. 3. Elute the products with RP-buffer B using an appropriate elution gradient (formic acid can be used instead of TFA if a LC/MS detector is employed). 4. Analyze the eluate with a UV–visible or fluorescence detector at a suitable wavelength. 5. In case further analysis of the samples is desired (e.g., SDS electrophoresis, mass spectrometry), collect the fractions containing the conjugate and use dialysis or ultrafiltration to change the buffer.

122

Mero et al.

6. Lyophilize the conjugate products and store at −20°C until use (see Note 8). 3.4.4. Size-Exclusion Chromatography

1. Equilibrate the size-exclusion column and the refractive index (RI) detector with elution buffer for 1  h at a flow rate of 1 mL/min. 2. Load 20  mL of the PEG–protein conjugates (0.1–0.5  mg/ mL protein). 3. Elute the product with the same elution buffer used to equilibrate the column in Step 1. 4. To analyze the eluate, the system can be connected to two channels: a UV–visible or fluorescence detector and a RI detector.

3.4.5. SDS-Polyacrylamide Gel Electrophoresis and PEG–Protein Detection

1. Run the PEG–protein conjugate sample on an SDS-PAGE gel according to the electrophoresis apparatus manufacturer’s instructions. 2. After separating the PEG–protein conjugates by electrophoresis, soak the resulting SDS-PAGE gel in 20 mL of perchloric acid (0.1 M) for 15 min. 3. Add 5 mL of BaCl2 solution and 2 mL of iodine solution. The brown-stained PEG bands should appear within a few minutes. 4. After 10–15 min, replace the staining solution with H2O and incubate the gel for another 15 min. 5. The iodine-stained gel can also be further stained with Coomassie blue for the detection of proteins.

3.4.6. Mass Spectrometry of PEG–Protein Conjugates

1. Dissolve the PEG–protein conjugate sample (20 mg) in a 0.1% (v/v) TFA aqueous solution (5–10 mL). 2. Mix a saturated matrix solution with the PEG–protein sample solution in the ratio of 1:1 (v/v). 3. Load the mixture (10–20  mL) onto the sample plate and perform the MS analysis procedure after solvent evaporation.

3.4.7. Determination of PEGylation Sites for Proteins Modified with a Polydisperse PEG

1. Separately dissolve the native and PEGylated proteins (200 mg) in Tris-Gdn buffer (pH 9.0) at a final protein concentration of 1 mg/mL. 2. Add tris (2-carboxyethyl) phosphine to the protein solution at a final concentration of 5 mM. Incubate the reaction mixture for 1 h at 37°C. 3. Add iodoacetamide (25 mM) to the reduced protein solution and incubate the reaction mixture for 30 min at 37°C in the dark. 4. Purify the protein samples by RP-HPLC using a C18 column. Dry the collected fractions.

Covalent Conjugation of Poly(Ethylene Glycol) to Proteins and Peptides

123

5. Dissolve the reduced and S-carboxamidomethylated samples of native and PEGylated protein in 8  M urea. Next, dilute the samples further in phosphate buffer to obtain a final protein concentration of 0.8 mg/mL and a final urea concentration of 0.8 M. 6. Add trypsin at an enzyme/substrate (E/S) ratio of 1:50 (w/w) and let the proteolysis reaction proceed at 37°C overnight (see Note 10). 7. Fractionate (e.g., use an analytical C4 or C18 column) both the native and PEGylated protein digestion mixtures by RP-HPLC. Collect the products by monitoring the eluate by UV absorbance (l = 214 nm). 8. Compare the elution patterns of the peptides obtained from the modified and native digests. The identity of the peptides that are missing in the PEGylated protein digest can be established by analysis of the corresponding peaks in the non-PEGylated digest. For this purpose, mass spectrometry is employed. 9. The elution pattern obtained from the modified (PEGylated) protein may sometimes show new peaks corresponding to the PEGylated peptides. Their identity can be revealed by MALDI-TOF mass spectrometry, even though this may not be an easy task (due to the polydispersity of PEG). 3.4.8. Determination of PEGylation Sites for Proteins Modified with a Monodisperse PEG

Follow Steps 1–6 in Subheading 3.4.7 above, and then continue with the following procedure: 1. Desalt the native and PEGylated protein digestion mixtures using a PepClean™ C-18 spin column and analyze the products directly by ESI-MS. 2. For both the native protein and PEGylated protein samples, identify all the resulting peptide fragments. Some of the peptides obtained in the PEGylated sample may show an increase in mass corresponding to the conjugation of a single chain of polymer. 3. If more than one available site for conjugation is present within the same fragment, a tandem mass spectrometry (MS/ MS) analysis is performed to determine which particular amino acid is modified.

3.4.9. Degree of Protein Modification by the Habeeb Assay

1. Prepare the native (unmodified) protein sample and the PEGylated derivative at the same protein concentration (0.2– 0.8 mg/mL) in phosphate buffer, pH 7.2. 2. Prepare the TNBS reagent and perform the assay in test tubes (in duplicate) at room temperature according to Table 5. 3. Incubate all the samples in a water bath at 40°C for 2 h, and then add 250 mL of 10% (w/v) SDS and 125 mL of 1N HCl. 4. Read the absorbance of the solution at l = 335  nm using a spectrophotometer.

124

Mero et al.

Table 5 Preparation of test solutions for the Habeeb assay Blank

Native protein

Pegylated protein

250 mL phosphate pH 7.2

250 mL of protein sample

250 mL of PEGylated sample

250 mL bicarbonate pH 8.5

250 mL bicarbonate pH 8.5

250 mL bicarbonate pH 8.5

250 mL TNBS

250 mL TNBS

250 mL TNBS

Table 6 Preparation of test solutions for the indirect Ellman’s assay Blank

Native protein

Pegylated protein

50 mL of Ellman’s reagent

50 mL of Ellman’s reagent

50 mL of Ellman’s reagent

1 mL of phosphate pH 7.2

970 mL of phosphate pH 7.2

970 mL of phosphate pH 7.2

30 mL of protein sample

30 mL of PEGylated sample

The degree (%) of amine substitution is calculated as follows:

% Substitution = [1 - (AP - AB)/(AN - AB)] ´ 100%

AP = Absorbance of the PEGylated protein AB = Absorbance of the blank AN = Absorbance of the native protein. 3.4.10. Degree of Cysteine Modification by the Indirect Ellman’s Assay

1. Prepare the native (unmodified) protein sample and the PEGylated derivative at the same protein concentration (0.2–0.8 mg/mL) in phosphate buffer, pH 7.2. 2. Prepare the Ellman’s reagent and perform the assay in test tubes (in duplicate) at room temperature according to Table 6. 3. Incubate all the samples for 15 min. 4. Read the absorbance spectrophotometer.

at

l = 412 

nm

using

a

The percentage of free SH groups is calculated by applying the following formula:

% Free thiol groups = [(AP - AB)/(AN - AB)] ´ 100

AP = Absorbance of the PEGylated sample AB = Absorbance of the blank AN = Absorbance of the native protein.

Covalent Conjugation of Poly(Ethylene Glycol) to Proteins and Peptides

125

Table 7 Preparation of test solutions for the iodine assay Blank

Sample

525 mL of Milli-Q water

500 mL of Milli-Q water

250 mL of BaCl2 solution

250 mL of BaCl2 solution

250 mL of iodine solution

250 mL of iodine solution 25 mL of PEG solution

3.4.11. Qualitative Test for the Presence of PEG by the Iodine Assay

A rapid, qualitative analysis of the total PEG content in a sample can be performed as follows: 1. To a clean tube, add 975 mL of deionized water, 250 mL of BaCl2 solution, and 250 mL of iodine solution. 2. To the above mixture, add 25 mL of the PEG–protein conjugate sample solution. 3. The test is positive if the final mixture forms a dark precipitate, or if it shows increased absorbance at l = 535 nm.

3.4.12. Quantitative Test for the Amount of PEG by the Iodine Assay

A quantitative analysis of the total PEG content in a sample can be performed as follows (see Note 11): 1. Prepare the blank, PEG standard solutions (0.2–0.5  mg/ mL PEG) and unknown sample solutions according to Table 7. 2. Incubate the solutions for 15 min and then read the absorbance at l = 535 nm. 3. Generate a calibration curve by plotting the measured absorbance values (535 nm) versus the known concentration values of the PEG standards. 4. The amount of PEG present in the unknown sample solutions can be determined from comparison of the measured absorbance values against the standard curve generated in Step 3.

4. Notes 1. A limitation of the use of PEG is its hygroscopicity: If not stored under dry conditions, activated PEGs are easily hydrolyzed. It is useful to always verify the degree of activation of new batches of activated PEGs that have been obtained

126

Mero et al.

commercially. Activation values between 70 and 90% are acceptable, but they must be taken into consideration when determining the excess amount of PEG needed for a desired conversion yield. To minimize the levels of deactivation due to hydrolysis, store activated PEG reagents at −20°C under nitrogen and warm the bottle to room temperature before opening. 2. A major limitation of PEG aldehyde derivatives is their susceptibility to air oxidation. Low-temperature storage under an inert atmosphere is mandatory, even if such conditions may not always be effective. 3. When PEG-OPSS is used, careful attention should be paid to avoid the presence of any thiol-reducing agents in all steps of conjugate preparation and purification. 4. When PEG-Mal is used, avoid pH conditions above 7.5 since at higher pH, reaction with primary amine groups can also take place (although at a slower rate compared with free thiol groups). 5. Different aqueous buffers may be employed (e.g., phosphate, borate, HEPES, etc.), but do not use tris (hydroxymethyl) aminomethane (Tris) or any other primary amine-containing buffer components because they will compete with proteins in the PEG coupling reaction. 6. Dialysis is effective if the outside buffer is often changed and the dialysis volume is 500- to 1,000-fold greater than the volume of the sample. 7. The mobile phase in CEX can be a phosphate, acetate, or citrate buffer solution. Specific buffers are provided in Subheading  2.3 as examples, but the optimal buffer to use needs to be considered case-by-case depending on the particular protein studied. 8. Some proteins are also stable at room temperature provided that maintenance of sterility is guaranteed. 9. For RP-HPLC buffer B, CH3CN is typically used, but in some cases a mixture of CH3CN/MeOH or only MeOH may also be suitable. 10. Different enzymes may be used to digest the protein samples; trypsin, chymotrypsin, and V8-protease are the most commonly employed. 11. Iodine assays for the quantitative determination of PEG must be performed after purification of the conjugates from free, unbounded PEG molecules.

Covalent Conjugation of Poly(Ethylene Glycol) to Proteins and Peptides

127

References 1. Abuchoswki, A., Van, E., Palczuk, N.C., et al. (1977) Alteration of immunological properties of bovine serum albumin by covalent attachment of polyethylene glycol. J. Biol. Chem. 252, 3578–3581. 2. Jevševar, S., Kunstelj, M., Gaberc Porekar, V. (2010) PEGylation of therapeutic proteins. J. Biotechnol. 5, 113–128. 3. Mero, A., Veronese, F.M. (2008) The impact of PEGylation on biological therapies Biodrugs 22, 315–329. 4. Zalipsky, S. (1995) Functionalized poly (ethylene glycol) for preparation of biologically relevant conjugates. Bioconjug. Chem. 6, 150–165. 5. Veronese, F.M. (2001) Peptide and protein PEGylation: a review of problems and solutions. Biomaterials 22, 405–417. 6. Snyder, S. L. and Sobocinski, P. Z. (1975) An improved 2,4,6,-trinitrobenzenesulfonic acid method for the determination of amines. Anal. Biochem. 64, 284–288. 7. Riddles, P.W., Blakeley, R. L., and Zarner, B. (1979) Ellman’s reagent: 5,5’-dithiobis(2nitrobenzoic acid) a reexaminatio. Anal. Biochem. 94, 75–81. 8. www.laysanbio.com 9. Hanton, S. D. (2001) Mass spectrometry of polymers and polymer surfaces. Chem. Rev. 101, 527–570. 10. Vestling, M. M., Murphy, C. M., Fenselau, C., Dedinas, J., Doleman, M. S., Harrsch, P. B., Kutny, R., Ladd, D. L., and Olsen, M. A. (1992) Techniques in Protein Chemistry III, Academic Press, New York, pp. 477–485. 11. Vestling, M. M., Murphy, C. M., Keller, D. A., Fenselau, C., Dedinas, J., Ladd, D. L., and Olsen, M. A. (1993) A strategy for character­ ization of polyethylene glycol-derivatized proteins: a mass spectrometric analysis of the attachment sites in polyethylene glycol-derivatized superoxide dismutase. Drug Metab. Dispos. 21, 911–917. 12. Montaudo, G., Samperi, F., Montaudo, M. S. (2006) Characterization of synthetic polymers by MALDI-MS. Prog. Polym. Sci. 10, 1016–1020. 13. Zhang H., Zhang, J., Luo, Y., Wilson, J., and Miller, K. (2007) Mass spectrometric analyses of potential impurities in 20 kDa monomethoxy poly (ethylene glycol)-propionaldehyde (PEG aldehyde) raw material. AAPS Annual Meeting & Exposition. 14. Sivakolundu, S. G., and Mabrouk, P. A. (2003) Proton NMR study of chemically modified

horse heart ferricytochrome C confirms the presence of histidine and lysine-ligated conformers in 30% acetonitrile solution. J. Inorg. Biochem. 94, 381–385. 15. Orsatti, L. and Veronese, F. M. (1999) An unusual coupling of poly(ethylene glycol) to tyrosine residues in epidermal growth factor. J. Bioact. Compat. Pol. 14, 429–436. 16. Wylie, D. C., Voloch, M., Lee, S., Liu, Y. H., Cannon-Carlson, S., Cutler, C., Pramanik, B. (2001) Carboxyalkylated histidine is a pHdependent product of PEGylation with SC-PEG. Pharm Res. 18, 1354–1360. 17. Kinstler, O. B., Brems, D. N., Lauren, S. L., Paige, A. G., Hamburger, J. B. and Treuheit, M. J. (1996) Charecterization and stability of N-Terminally PEGylated rhG-CSF. Pharm. Res. 13, 996–1002. 18. Lee, H., Jang, I. H., Ryu, S. H., and Pack, T.G. (2003) N-Terminal site-specific monoPEGylation of epidermal growth factor. Pharm. Res. 20, 818–825. 19. Kinstler, O., Molineux, G., Treuheit, M., Ladd, D. et  al. (2002) Mono-N-terminal poly(ethyleneglycol)-protein conjugates. Adv. Drug Deliv. Rev. 54, 477–485. 20. Arakawa, T., Prestrelski, S. J., Narhi, L. O., Boone, T. C., Kenney, W. C. (1993) Cysteine 17 of recombinant human granulocyte-colony stimulating factor is partially solvent-exposed. J. Protein Chem. 12, 525–531. 21. Colonna, C., Conti, B., Perugini, P., Pavanetto, F., Modena, T., Dorati, R., Iadarola, P., Genta, I. (2008) Site-directed PEGylation as successful approach to improve the enzyme replacement in the case of prolidase. Int. J. Pharm. 24, 230–237. 22. Xian-Hui, H., Pang-Chui, S., Li-Hui, X., and Siu-Cheung, T. (1999) Site-directed polyethylene glycol modification of trichosanthin: Effects on its biological activities, pharmacokinetics, and antigenicity. Life Sic. 64, 1163–1175. 23. Balan, S., Choi, J., Godwin, A., Teo, I., Laborde, C. M., Heidelberger, S., Zloh, M., Shaunak, S., and Brocchini, S. (2007) SiteSpecific PEGylation of Protein Disulfide Bonds Using a Three-Carbon Bridge. Bioconjug. Chem. 18, 61–76. 24. Zalipsky, S., and Meno-Rudolph, S. (1997) Hydrazide derivatives of polyethylene glycol and their bioconjugates. In Polyethylene glycol chemistry and biological applications, ACS sym­ posium series 680 (Harris, J. M., and Zalipsky, S. eds.), pp. 318–341.

128

Mero et al.

25. Gaertner, H. F., and Offord, R. E. (1996) Site-Specific Attachment of Functionalized Poly(ethylene glycol) to the Amino Terminus of Proteins. Bioconjug. Chem. 7, 38–44. 26. Baudys, M., Uchio, T., Mix, D., Wilson, D., and Kim, S. W. (1995) Physical stabilization of insulin by glycosylation. J. Pharm. Sci. 84, 28–33. 27. Esposito, P., Barbero, L., Caccia, P., Caliceti, P., D’Antonio, M., Piquet, G., Veronese, F. M. (2003) PEGylation of growth hormone-releasing hormone (GRF) analogues. Adv. Drug Del. Rev. 55, 1279–1291 28. Veronese, F. M., Mero, A., Caboi, F., Sergi, M., Marongiu, C., and Pasut, G. (2007) SiteSpecific PEGylation of G-CSF by Reversible Denaturation. Bioconjug. Chem. 18, 1824–1830. 29. Morpurgo, M., Monfardini, C., Hofland, L. J., Sergi, M., Orsolini, P., Dumont, J. M., and Veronese, F. M. (2002) Selective Alkylation and Acylation of a and e Amino Groups with PEG in a Somatostatin Analogue: Tailored Chemistry for Optimized Bioconjugates. Bioconjug. Chem. 13, 1238–1243. 30. Youn, Y. S., Na, D. H., Lee, K. C. (2007) High-yield production of biologically active mono-PEGylated salmon calcitonin by sitespecific PEGylation. J. Control Release 117, 371–379. 31. Chae, S. Y., J., C. H., Shin, H. J., Youn, Y. S., Lee, S., Lee, K.C. (2008) Preparation, characterization, and application of biotinylated and biotin-PEGylated glucagon-like peptide-1 analogues for enhanced oral delivery. Bioconjug. Chem. 19, 334–341. 32. Sato, H. (2002) Enzymatic procedure for sitespecific PEGylation. Adv. Drug Delivery. Rev. 54, 487–504. 33. Fontana, A., Spolaore, B., Mero, A., and Veronese, F. M. (2008) Site-specific modification and PEGylation of pharmaceutical proteins mediated by transglutaminase. Adv. Drug Delivery Rev. 60, 13–28. 34. Li, N., Ziegemeier, D., Bass, L., Wang, W. (2008) Quantitation of free polyethylene glycol in PEGylated protein conjugate by size exclusion HPLC with refractive index (RI) detection. J. Pharm Biomed Anal. 48, 1332–1338. 35. Sims, G. E., and Snape, T. J. (1980) A method for the estimation of polyethylene glycol in plasma protein fractions. Anal. Biochem. 107, 60–63. 36. Seely, J. E., Richey, C. W. (2001) Use of ionexchange chromatography and hydrophobic interaction chromatography in the preparation and recovery of polyethylene glycol-linked proteins. J. Chromatogr. A 908, 235–241.

37. Seely, J. E., Buckel, S. D., Green, P. D., Richey, C. W. (2005) Making site-specific PEGylation work. Biopharm Int. 18, 30–35. 38. Wang, Y. S., Youngster, S., Bausch, J., Zhang, R., McNemar, C., Wyss, D. F. (2000) Identification of the major positional isomer of pegylated interferon alpha-2b. Biochemistry 39, 10634–10640. 39. Fee, C. J., Van Alstine, J. M. (2006) PEGproteins: Reaction engineering and separation issues. Chem. Eng. Sci. 61, 924–939. 40. Pabst, T. M., Buckley, J. J., Ramasubramanyan, N., Hunter, A. K. (2007) Comparison of strong anion-exchangers for the purification of a PEGylated protein. J. Chromatogr. A 1147, 172–182. 41. Park, E. J., Lee, K. C, Na, D. H. (2009) Separation of positional isomers of monopoly(ethylene glycol)-modified octreotides by reversed-phase high-performance liquid chromatography. J Chromatogr. A 6, 7793–7797. 42. Clark, R., Olson, K., Fuh, G., Mariani, M., Mortensen, D., Teshima, G., Chang, S., Chu, H., Mukku, V., Canova-Davis, E., Somers, T., Cronin, M., Winkler, M., and Wells, J. A. (1996) Long-acting Growth Hormones Produced by Conjugation with Polyethylene Glycol. J. Biol. Chem. 271, 21969–21977. 43. Foser, S., Schacher, A., Weyer, K. A., Brugger, D., et al. (2003) Isolation, structural characterization, and antiviral activity of positional isomers of monopegylated interferon alpha-2a (PEGASYS). Protein Expr. Purif. 30, 78–87. 44. Gaberc-Porekar, V., Zore, I., Podobnik, B., Menart,V. (2008) Obstacles and pitfalls in the PEGylation of therapeutic proteins. Curr. Opin. Drug Discov. Devel. 11, 242–250. 45. Piedmonte, D. M., Treuheit, M. J. (2008) Formulation of Neulasta(R) (pegfilgrastim). Adv. Drug Del. Rev. 60, 50–58. 46. Fee, C. J (2007) Size comparison between proteins PEGylated with branched and linear poly(ethylene glycol) molecules. Biotechnol Bioeng. 98, 725–731. 47. Edwards, C. K., Martin, S. W., Seely, J., Kinstler, O. et al. (2003) Design of PEGylated soluble tumor necrosis factor receptor type I (PEG sTNF-RI) for chronic inflammatory diseases. Adv. Drug Deliv. Rev. 55, 1315–1336. 48. Molek, J. R., Zydney, A. L. (2006) Ultrafiltration characteristics of pegylated proteins. Biotechnol. Bioeng. 95, 474–482. 49. Kwon, B., Molek, J., Zydney, A. L. (2008) Ultrafiltration of PEGylated proteins: Fouling and concentration polarization effects. J. Memb. Sci. 319, 206–213.

Covalent Conjugation of Poly(Ethylene Glycol) to Proteins and Peptides 50. Habeeb, A. F. S. A. (1966) Determination of free amino groups in protein by trinitrobenzenesulphonic acid. Anal.Biochem. 14, 328–336. 51. Riddles, P. W., Blakeley, R. L., and Zarner, B. (1983) Reassessment of Ellman’s reagent. Methods Enzymol. 91, 49–60. 52. Jiskoot, W., Crommelin, D. (2005) Methods for Structural Analysis of Protein Pharmace­ uticals Biotechnology: Pharmaceutical Aspects. American Assoc. of Pharm. Scientists, Springer, New York. 53. www.piercenet.com 54. Kusterle, M., Jevsevar, S., Gaberc-Porekar, V. (2008) Size of Pegylated Protein Conjugates Studied by Various Methods. Acta Chim. Slov. 55, 594–601. 55. Piedmonte, D. M., Treuheit, M. J. (2008) Formulation of Neulasta(R) (pegfilgrastim). Adv. Drug Del. Rev. 60, 50–58. 56. Lee, K., Moon, S. C., Park, M. O., Lee, J. T., Na, D. H., Yoo, S. D., et al. (1999) Isolation, characterizasion, and stability of positional isomers of mono-PEGylated salmon calcitonins. Pharm. Res. 16, 813–818. 57. Manjula, B. N., Tsai, A., Upadhya, R., Perumalsamy, K., Smith, P. K., Malavalli, A., Vandegriff, K. R., Winslow, M., Intaglietta, M., Prabhakaran, M., Friedman, J. M., and Acharya A. S. (2003) Site-Specific PEGylation of Hemoglobin at Cys-93(b): Correlation between the Colligative Properties of the PEGylated Protein and the Length of the Conjugated PEG Chain. Bioconjug. Chem. 14, 464–472. 58. Fee, C. J., Van Alstine, J. M. (2004) Prediction of the viscosity radius and the size exclusion chromatography behavior of PEGylated proteins. Bioconjug. Chem. 15, 1304–1313. 59. Kurfurst, M. M. (1992) Detection and Molecular-Weight Determination of Polyethylene Glycol-Modified Hirudin by Staining After Sodium Dodecyl-Sulfate Polyacrylamide-Gel Electrophoresis. Anal. Biochem. 200, 244–248. 60. Caccia, D., Ronda, L., Frassi, R., Perrella, M., Del Bavero, E., Bruno, S., Pioselli, B., Abbruzzetti, S., Viappiani, C., and Mozzarelli A. (2009) PEGylation Promotes Hemoglobin

129

Tetramer Dissociation. Bioconjug. Chem. 20, 1356–1366. 61. Caserman, S., Kusterle, M., Kunstelj, M., Milunovic, T. et al. (2009) Correlations between in  vitro potency of polyethylene glycol-protein conjugates and their chromatographic behaviour. Anal. Biochem. 389, 27–31. 62. Cindric, M., Cepo, T., Galic, N., BukvicKrajacic, M. et al. (2007) Structural characterization of PEGylated rHuG-CSF and location of PEG attachment sites. J. Pharm. Biomed. Anal. 44, 388–395. 63. Mero, A., Spolaore, B., Veronese, F. M., and Fontana, A. (2009) TransglutaminaseMediated PEGylation of Proteins: Direct Identification by Mass Spectrometry Using a Novel Monodisperse PEG. Bioconjug. Chem. 20, 384–389. 64. Sergi, M., Caboi, F., Maullu, C., Orsini, G., and Tonon, G. (2009) Enzymatic techniques for PEGylation of biopharmaceuticals. In PEGylated Protein Drugs: Basic Science and Clinical Applications (Milestones in Drug Therapy) (Veronese, F.M., ed.) Birkhauser Verlag, Boston, MA, pp. 75–88. 65. Basu, A., Yang, K., Wang, M., Liu, S., et  al. (2006) Structure-function engineering of interferon-beta-1b for improving stability, solubility, potency, immunogenicity, and pharmacokinetic properties by site-selective mono-PEGylation. Bioconjug. Chem. 17, 618–30. 66. Stoscheck, C. M. (1990) Quantification of protein. Methods Enzymol. 182, 50–69. 67. Yu, P., Zheng, C., Chen, J., Zhang, et  al. (2007) Investigation on PEGylation strategy of recombinant human interleukin-1 receptor antagonist. Bioorg. Med. Chem. 15, 5396–405. 68. Monfardini, C., Schiavon, O., Caliceti, P., Morpurgo, M., Harris, J. M., and Veronese, F. M. (1995) A branched mono­methoxypoly (ethylen glicol) for protein modification. Bioconjug. Chem. 6, 62–69. 69. Bailon, P., Palleroni, A., Schaffer, C. A., Spence, C. L. et al. (2001) Rational design of a potent, long-lasting form of interferon: a 40 kDa branched polyethylene glycol-conjugated interferon alpha-2a for the treatment of hepatitis C. Bioconjug. Chem. 12, 195–202.

Chapter 9 Extending the Scope of Site-Specific Cysteine Bioconjugation by Appending a Prelabeled Cysteine Tag to Proteins Using Protein Trans-Splicing Tulika Dhar, Thomas Kurpiers, and Henning D. Mootz Abstract Incorporating synthetic probes site-specifically into proteins is of central interest in several areas of biotechnology and protein chemistry. Bioconjugation techniques provide a simple and effective means of chemically modifying a protein. In particular, covalent chemical modifications of cysteine residues belong to one of the most important reactions due to the unique reactivity of its thiol moiety and the relatively low abundance of this amino acid in proteins. However, such types of modifications cannot be performed in a regioselective fashion when one or more additional cysteines are present. To address this limitation, we have developed an approach where a short cysteine-containing tag (Cys-Tag) fused to one part of a split intein and modified at its sulfhydryl group can be used to label proteins by trans-splicing with a protein of interest (POI) fused to the other half of the split intein. In this way, it is possible to selectively label a protein containing multiple cysteines. The artificially split Mycobacterium xenopi GyrA intein and the Synechocystis sp. DnaB intein were highly suitable for this purpose and were successfully used for the labeling of several proteins. This approach enables a simple route for labeling proteins by site-specific cysteine bioconjugation with any one of several commercially available cysteine-modifying probes. Key words: Cys-Tag, Split intein, Protein trans-splicing, Protein modification, Bioconjugation

1. Introduction Inteins are internal polypeptide sequences of proteins that can remove themselves autocatalytically without utilizing external co-factors or energy in a process-termed protein splicing (1–3). Besides a few naturally occurring split inteins, many cis-splicing inteins can be artificially split into two parts. These complementary intein halves, IntN and IntC, require first the association and folding of both halves to facilitate protein trans-splicing activity (1–7). Similar to a regular intein, they excise themselves out by breaking two peptide bonds and concomitantly join their ­respective Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_9, © Springer Science+Business Media, LLC 2011

131

132

Dhar, Kurpiers, and Mootz

flanking extein sequences (ExtN and ExtC) by a new peptide bond. This posttranslational reaction has successfully been used for several in vitro applications, such as protein semisynthesis (7, 8), modulation of extein sequences (9), and segmental isotopic labeling (10). Site-specific modification of proteins is an attractive tool for studying protein structure and function, and also for endowing them with new properties. However, each technique has certain advantages and limitations. Therefore, new techniques are required to expand the scope of accessible proteins and modifications. Split inteins combine general and easy applicability and variability with cost effectiveness. Figure 1 shows the concept of adding a CysTag to a protein of interest (POI) using split inteins. This approach not only selectively labels the protein, but the intein excision makes the reaction almost traceless, as well (11, 12). It also has the advantage of leaving other essential cysteines unaffected. In addition, a large number of synthetic probes, e.g., fluorophores, biotin, and PEG coupled to well-established ­functional groups for bioconjugation (like haloacetamides and maleimides) are commercially available from various suppliers. Furthermore, this approach requires only very low concentrations of the POI–intein fusion protein (in the low micromolar range). The trans-splicing reaction is highly selective due to the required association of both

Fig. 1. General scheme for the two-step approach of site-selectively labeling a protein of interest by a combination of cysteine modification and protein trans-splicing. (a) Covalent chemical modification of a cysteine in the Cys-Tag fused to the IntC auxiliary protein (RG = reactive group). (b) Appending the prelabeled Cys-Tag to the protein by protein transsplicing. A native peptide bond is formed between the protein and the Cys-Tag.

Extending the Scope of Site-Specific Cysteine Bioconjugation

133

complementary split intein fragments (in fusion with the POI), allowing the technique to be performed on crude proteins in cell lysates. A prerequisite for this method is that the split intein halfattached to the Cys-Tag must itself be completely free of cysteines. Most commonly known inteins harbor a catalytic cysteine at each splice junction that is required for the protein splicing mechanism to function. In contrast, the artificially split Ssp DnaB intein from Synechocystis sp. has a serine at the (+1) position of the C-terminal intein fragment (11). Similarly, the C-terminal nucleophile in the case of the artificially split Mxe GyrA intein from Mycobacterium tuberculosis is a threonine, thereby rendering it suitable for our cysteine-labeling approach (12) (see Note 1). Inteins were artificially split for several applications before the discovery of a naturally split intein. Many inteins are bi-functional elements consisting of a protein splicing domain and an endonuclease domain responsible for its mobility (intein homing). It was reported that the 275-amino acid endonuclease region could be completely deleted from the Ssp DnaB intein without loss of the protein splicing activity, resulting in a functional mini-intein of 154 amino acids. Furthermore, the Ssp DnaB mini-intein could be split into two halves that reconstitute splicing activity when co-expressed in Escherichia coli (13). Our lab was then able to demonstrate that the separately expressed and purified halves of the intein are active in trans-splicing under native conditions, and are therefore highly suitable for in vitro applications (14). The Mxe GyrA intein is a 198-amino acid native mini-intein consisting of a short linker region in place of the endonuclease domain (15). This intein was chosen and artificially split in our lab on the basis that – being a native mini-intein – it would prove beneficial in terms of solubility and efficiency for in vitro reconstitution. In fact, it proved superior to the Ssp DnaB intein and is our favored intein for the Cys-Tag approach (12) (see Notes 2 and 3). This approach may further be extended to other inteins and tags (see Notes 1 and 4). It can also be used for labeling an internal position of the POI if the POI is split at an appropriate site (16). A second chemical group could be incorporated in the protein after the trans-splicing. This technique can also be applied suitably to obtain N-terminally labeled proteins, as demonstrated for the Psp-GBD Pol intein (16). The following protocol is written based on the use of fusion proteins and conditions recently reported by our laboratory (11, 12). Specifically, it describes the Cys-Tag labeling approach using the split Mxe GyrA intein. However, the methodologies presented in this chapter can be used as a general guideline for devising strategies to append a labeled Cys-Tag to any POI under different conditions. The example we report in detail herein is the ­modification of the nonribosomal peptide synthetase TycA from Bacillus brevis, as illustrated in Fig. 2.

134

Dhar, Kurpiers, and Mootz

Fig. 2. Illustration of a specific example of the Cys-Tag bioconjugation strategy presented in this chapter. (a) Alkylation of the Cys-Tag in auxiliary protein MBP-IntC-Cys-Tag-His6 with 5¢-iodoacetamide fluorescein. (b) Transfer of the Cys-Tag to the protein TycA by protein trans-splicing. Note that in the fusion constructs, the MBP and FKBP moieties serve to assist in maintaining protein solubility and correct protein folding.

2. Materials 2.1. Expression Plasmid Construction ( see Figs. 2 and 3)

1. Plasmid pTK130 (encoding Strep-tag II-TycA-GyrAN-FKBP; protein 1) (12). 2. Plasmid pTK120 (encoding MBP-GyrAC-Cys-Tag-His6; protein 2) (12) (see Note 5). 3. Restriction enzymes: EcoRI, NheI (Fermentas).

2.2. Protein Expression and Purification

1. Escherichia coli BL21 Gold (DE3) cells (Stratagene). 2. LB medium: 5 g/l NaCl, 10 g/l tryptone, 5 g/l yeast extract, pH 7.0. 3. Ampicillin stock solution (50 mg/ml), filter sterilize. 4. Isopropyl-b-thiogalactoside (IPTG) stock solution (400 mM), filter sterilize. 5. Strep-Tactin® affinity chromatography resin. 6. Strep-Tactin® Buffer W: 100 mM Tris–HCl, 150 mM NaCl, 1 mM EDTA, pH 8. 7. Strep-Tactin® Buffer E: 100 mM Tris–HCl, 150 mM NaCl, 1 mM EDTA, 2.5 mM desthiobiotin, pH 8. 8. Amylose affinity chromatography resin.

Extending the Scope of Site-Specific Cysteine Bioconjugation

135

Fig. 3. Schematic representation of the cloning of any protein of interest (POI) as a fusion construct with the IntN auxiliary protein.

9. Amylose Buffer: 20  mM Tris–HCl, 200  mM NaCl, 1  mM EDTA, pH 7.4. 10. Amylose Elution Buffer: 20 mM Tris–HCl, 200 mM NaCl, 1 mM EDTA, 10 mM maltose, pH 7.4. 11. Ni2+-NTA immobilized metal affinity chromatography resin. 12. Ni2+-NTA Wash Buffer A: 50 mM Tris–HCl, 300 mM NaCl, pH 8.0. 13. Ni2+-NTA Wash Buffer B: 50 mM Tris–HCl, 300 mM NaCl, 250 mM imidazole, pH 8.0. 2.3. Cysteine Labeling

1. Dilution buffer: 50 mM phosphate buffer (pH 7.2) containing 150 mM NaCl. 2. 5-(Iodoacetamido)fluorescein (5-IAF) stock solution (10 mM): Dissolve 1 mg of 5-IAF in 20 ml of DMF, and then dilute with 174.1  ml of dilution buffer to get a final stock solution of 10 mM (see Note 6). 3. Dithiothreitol (DTT) stock solution (10  mM): Dissolve 1.54 mg of DTT in 1 ml of water to give a stock solution of 10 mM (freshly prepared). 4. Hi-Trap™ Sephadex™ G-25 Superfine desalting column (GE Healthcare).

2.4. Protein ­Trans-splicing

1. Splice Buffer: 50  mM Tris–HCl, 300  mM NaCl, 1  mM EDTA, pH 7.0. 2. 4× SDS-loading buffer: 500 mM Tris–HCl, pH 6.8, 8% (w/v) SDS, 40% (v/v) glycerol, 20% (v/v) b-mercaptoethanol, 4 g/l bromophenol blue.

136

Dhar, Kurpiers, and Mootz

3. Methods The Cys-Tag labeling approach can essentially be used for selectively modifying almost any POI. In our hands, this method has allowed us to successfully carry out the modification of several important proteins, such as human growth hormone (hGH), the nonribosomal peptide synthetase tyrocidine A (TycA), beta-lactamase, etc., with all of them containing one or more functionally or structurally important cysteine residues (11, 12). However, certain points are to be kept in mind regarding the nature of the intein. Artificially split inteins tend to behave differently than their intact parent proteins. When artificially split protein fragments are individually expressed, they often display a higher tendency to misfold or aggregate; this could be due to the exposure of hydrophobic patches, which in some cases can result in insolubility. On a similar note, many artificially split inteins require a refolding step under denaturing conditions to become active in protein trans-splicing, thus preventing their application to proteins that cannot be refolded. However, this is not the case for the above-mentioned Ssp DnaB and Mxe GyrA inteins, which can be purified from and reconstituted under native conditions. Another consideration to bear in mind is the nature of the amino acids directly flanking the intein. These amino acids can play a significant role in influencing the efficiency of the splicing reaction. In order to prevent distortions caused by their close proximity to the intein active site, it is generally recommended to keep three to five native residues on each side of the intein when inserting it into a foreign protein sequence. The most critical flanking residue is usually the (−1) amino acid located immediately upstream of the intein. The (+1) catalytic residue, a cysteine, serine or threonine, is required for the protein splicing reaction and is retained in the splice product. To generate the two halves of the Mxe GyrA intein, we split the encoding gene between positions for Arg119 and Gly120 on the DNA level, thereby creating a 119-amino acid IntN fragment and a 79-amino acid IntC fragment. The first two adjacent residues of the natural N-terminal extein sequence (positions −1 and −2) and the first three residues of the natural C-terminal extein sequence (positions +1 to +3) were retained in our constructs (plasmids pTK130 and pTK120, respectively). The resulting IntC is free of any cysteine residues. Additionally, to prevent the larger IntN fragment from potential aggregation and to minimize the exposure of hydrophobic regions which might lead to misfolding, a highly soluble protein like the phage protein gpD or FK506 binding protein (FKBP) domain was added at the C-terminus (12).

Extending the Scope of Site-Specific Cysteine Bioconjugation

3.1. Generating Int N Fusion Construct with POI

137

1. Amplify the gene encoding your POI by PCR with an N-terminal extension for the recognition sequence of NheI and a C-terminal extension for the recognition sequence of EcoRI (see Note 7). 2. Digest the PCR product and plasmid pTK130 with the restriction enzymes NheI and EcoRI. 3. Purify both fragments by agarose gel electrophoresis. 4. Ligate both fragments together with T4 DNA ligase to generate a complete plasmid encoding your POI (see Fig. 3).

3.2. Expression and Purification of Recombinant Protein (see Note 8)

1. Transform one E. coli BL21 (DE3) cell culture with the plasmid encoding the IntN (pTK130) fusion construct. Transform a second (i.e., separate) E. coli BL21 (DE3) cell culture with the plasmid encoding for the IntC (pTK120) fusion construct. 2. For each fusion construct, inoculate 600  ml of fresh LB medium containing 100  mg/ml ampicillin with 6  ml of an overnight seed culture. 3. Grow the cells at 37°C to an OD600 of 0.7. 4. Lower the culture temperature to 25°C and induce protein expression by adding 600 ml of IPTG stock solution to give a final concentration of 0.4 mM IPTG. 5. After 3–5 h, harvest the cells by centrifugation at 10,000 × g for 20 min at 4°C. 6. Resuspend the cells expressing the IntN fusion protein (protein 1) in cold Buffer W, and those expressing the IntC fusion protein (protein 2) in amylose buffer. 7. Either freeze the resuspended cells at −80°C or immediately proceed to disrupt the cells using an emulsifier (high-pressure homogenizer). Keep the cell suspension and the cell lysate on ice. 8. Centrifuge the cell lysate at 30,000 × g for 30 min at 4°C to separate the soluble fraction (supernatant) (see Note 2). 9. For the IntN fusion protein (protein 1), perform protein purification on a Strep-Tactin® affinity column (with gravity feed) according to the manufacturer’s instructions: Load the supernatant extract directly onto a Strep-Tactin® column pre-equilibrated with Buffer W. After the cell extract has completely entered the column, wash the column with three column volumes of Buffer W and then elute the fusion protein from the column using Buffer E. 10. For the IntC fusion protein, perform protein purification first on an amylose column, and then perform a second column purification using Ni2+-NTA as follows: Amylose column purification (gravity feed):  Fill a chromatography column with 5 ml of amylose resin and equilibrate with

138

Dhar, Kurpiers, and Mootz

Amylose Buffer. Load the supernatant extract and wash the column three times with Amylose Buffer. Elute the fusion protein from the column using Amylose Elution Buffer and pool the fractions containing the fusion protein. Perform a buffer exchange (by dialysis) to replace the Amylose Elution Buffer with Ni2+-NTA Buffer A for a second column purification using Ni2+-NTA. Ni2+-NTA column purification (gravity feed):  Fill a chromatography column with 2 ml of Ni2+-NTA resin and equilibrate with Buffer A. Load the protein sample (obtained after amylose column purification) and wash the column with 5 ml of Buffer A. Repeat the wash with 5 ml of Buffer A containing 5 mM imidazole. Wash twice more, once with Buffer A containing 20  mM imidazole and the last wash with Buffer A containing 40 mM imidazole. Finally, elute the fusion protein from the column using Buffer B. 11. Analyze the protein samples by SDS-PAGE: Mix 10 ml of protein sample with 3 ml of 4× SDS-loading dye, and load 10 ml onto the gel to check for purity. 12. Dialyze each of the proteins against splice buffer containing 2 mM DTT and 10% (v/v) glycerol. 13. Calculate the protein concentration of the dialyzed samples using the calculated molecular extinction coefficient at l = 280 nm. 14. Aliquot the purified fusion proteins into Eppendorf tubes, and flash-freeze in liquid nitrogen. Store the proteins at −80°C. 3.3. Cys-Tag Labeling

1. Reduce protein 2 by adding a tenfold molar excess of reducing agent (e.g., to a 100 ml aliquot of a 20 mM protein solution, add 2 ml from a 10 mM DTT stock solution). Incubate the mixture for 10 min at room temperature. 2. Add the labeling reagent (e.g., 5-IAF) at a 25-fold molar excess over protein 2 (e.g., add 5 ml from a 10  mM 5-IAF stock solution). 3. Incubate the mixture at 25°C for 2 h. 4. Quench the labeling reaction with 2–10 mM DTT. 5. Remove excess labeling and reducing agent by gel filtration (optional).

3.4. Protein Transsplicing Reaction

1. Add protein 1 and labeled protein 2 together in equimolar concentrations in a 1.5-ml Eppendorf tube along with 2 mM DTT. Adjust the volume with splice buffer. We suggest using a concentration of 10 mM for each protein (adjust according to the eluted protein concentrations, or perform further concentration if desired).

Extending the Scope of Site-Specific Cysteine Bioconjugation

139

Fig. 4. SDS-PAGE analysis of the trans-splicing reaction of the Cys-Tag strategy described in this chapter. (a) Analysis of the products of the protein trans-splicing reaction between protein 1 and protein 2 on a Coomassie brilliant blue-stained gel (left ) and a UV-illuminated gel (right ). (b) Analysis of the labeled splice product obtained from the reaction mixture after performing two consecutive column chromatography purification procedures. Left: Coomassie-stained gel. Right : UV-illuminated gel. (Lane 1: flow-through of the Ni2+-NTA column; Lane 2: elution fraction of the Ni2+-NTA column; Lane 3: flow-through of the Strep-Tactin® column; Lane 4: elution fraction of the Strep-Tactin® column.).

2. Incubate the reaction mixture at 25°C and remove 10-ml ­aliquots at different time points to monitor the progress of the reaction by SDS-PAGE (see Note 9). 3. Stop the reaction by mixing each aliquot with 3 ml of 4× SDSPAGE loading buffer. 4. Boil the samples for 10 min and load 10 ml onto the SDSPAGE gel. 5. Observe the gel on a UV transilluminator. 6. Stain the gel with Coomassie brilliant blue (see Fig. 4a and Note 10). 7. Scan the gel to determine the relative intensities of the protein bands (see Note 11). 3.5. Purification of Labeled Splice Product (see Fig. 4b and Note 12)

1. Perform the splice reaction as described Subheading 3.4, step 1 for approximately 4 h.

above

in

2. Apply the entire reaction mixture onto a Ni2+-NTA column and perform column purification as described earlier in Subheading 3.2, step 10. 3. Collect the flow through and apply the protein sample onto a Strep®-Tactin column and again perform column purification as described earlier in Subheading 3.2, step 9.

4. Notes 1. There are other known inteins that in principle could be used similarly. For example, the Tli-Pol2 intein has a threonine as the +1 nucleophile (17), whereas the Psp-GBD Pol intein

140

Dhar, Kurpiers, and Mootz

(16) and the Ssp GyrB intein (18) have a serine at this position. 2. A comparison between the split Ssp DnaB and Mxe GyrA inteins has revealed that GyrA has superior properties for protein trans-splicing applications. The product yield with GyrA is up to 70% (as revealed by densitometric analysis), in contrast to only a 40–50% yield with the Ssp DnaB intein, although the product formation is faster in the case of Ssp DnaB. Furthermore, the Mxe GyrA intein shows better solubility and has worked for several difficult proteins expressed in our lab. One of these, human growth hormone (hGH), contains four cysteine residues involved in disulfide bridges. hGH is generally difficult to express in E. coli and is typically found in the inclusion bodies when expressed as a hGH-IntN fusion. In addition, an attempt to refold the fusion protein by renaturation from 8 M urea proved unsuccessful. In order to circumvent this problem, the fusion protein was solubilized with 8  M urea, purified under these denaturing conditions, and then the complementary intein fragment (IntC-CysTag) was added to the 8  M urea solution before dialysis. Next, the reaction mixture was dialyzed against splice buffer at 4°C and then further incubated at 25°C for 2 h. The expected splice product, hGH-CysTag, was formed in high yields and remained fully in solution. Thus, a poorly soluble protein could be successfully labeled using a modified version of our approach with the split GyrA intein, which might be applicable to other such types of proteins as well (11, 12). 3. The two intein systems are orthogonal, and no product formation was seen when the constructs were incubated in either of the two heterologous combinations (12). 4. We have tested the suitability of this approach with several fluorophores, as well as by labeling with polyethylene glycol (PEG) moieties (12). 5. The Cys-Tag can be a short peptide tag as described here, or it can be part of a larger protein domain. The exact sequence of the attached Cys-Tag is variable, and at present has not been fully optimized. In this protocol, we have used a short peptide sequence in protein 2 (T[+1]EAGSCS), but in principle it could be shortened to even just two amino acids (i.e., the +1 nucleophile and the cysteine). 6. Fluorescein must be protected from light exposure in order to avoid photobleaching. 7. In this cloning scheme, if NheI and EcoRI cut within your POI then this strategy cannot be applied. Other techniques such as restriction-free cloning or the use of modified vectors to circumvent these sites may be applied instead.

Extending the Scope of Site-Specific Cysteine Bioconjugation

141

8. The optimum expression and purification conditions must be adjusted according to the specific POI being studied. 9. The splicing reaction can be carried out over a wide range of temperatures. We have found that the highest yields are obtained between 20 and 30°C, and that significant activity is still observed at temperatures as low as 12°C. Even at 4°C, a high yield is observed if the reaction is allowed to proceed for more than 24 h (12). 10. During the splice reaction with the Mxe GyrA intein, an additional protein band will be observed that has a higher molecular weight than both starting proteins. This band appears almost immediately after both proteins are mixed but disappears completely after 2  h. This band corresponds to the branched intermediate of the splicing reaction (12). 11. The product yield of the trans-splicing reaction is unaffected by the modification with the Cys-Tag label when compared with the unmodified product. 12. In the above-mentioned splice reaction, the splice product (obtained from Subheading  3.4) can be selectively purified using two types of affinity chromatography purification since it is the only product in the reaction mixture that contains both the Strep-tag II® and polyhistidine (6×His) affinity tags (see Fig. 4b). Conversely, this strategy can also be applied to remove other components and purify only the splice product if the splice product is the only species in the reaction mixture that contains no affinity tags.

Acknowledgments The author’s would like to thank all members of the Mootz lab for helpful discussions. T.D. acknowledges a Ph.D. stipend from the International Max Planck Research School in Chemical Biology. Funding for this work was provided by the DFG and the Fonds der Chemischen Industrie. References 1. Noren, C. J., Wang, J., and Perler, F. B. (2000) Dissecting the Chemistry of Protein Splicing and Its Applications, Angew Chem Int Ed Engl 39, 450–466. 2. Paulus, H. (2000) Protein splicing and related forms of protein autoprocessing, Annu Rev Biochem 69, 447–496. 3. Gogarten, J. P., Senejani, A. G., Zhaxybayeva, O., Olendzenski, L., and Hilario, E. (2002)

Inteins: structure, function, and evolution, Annu Rev Microbiol 56, 263–287. 4. Wu, H., Hu, Z., and Liu, X. Q. (1998) Protein trans-splicing by a split intein encoded in a split DnaE gene of Synechocystis sp. PCC6803, Proc Natl Acad Sci USA 95, 9226–9231. 5. Mills, K. V., Lew, B. M., Jiang, S., and Paulus, H. (1998) Protein splicing in trans by purified Nand C-terminal fragments of the Mycobacterium

142

Dhar, Kurpiers, and Mootz

tuberculosis RecA intein, Proc Natl Acad Sci USA 95, 3543–3548. 6. Evans, T. C., and Xu, M.-Q. (2002) Mechanistic and Kinetic Considerations of Protein Splicing, Chemical Reviews 102, 4869–4884. 7. Mootz, H. D. (2009) Split inteins as versatile tools for protein semisynthesis, Chembiochem 10, 2579–2589. 8. Muralidharan, V., and Muir, T. W. (2006) Protein ligation: an enabling technology for the biophysical analysis of proteins, Nat Methods 3, 429–438. 9. Southworth, M. W., Adam, E., Panne, D., Byer, R., Kautz, R., and Perler, F. B. (1998) Control of protein splicing by intein fragment reassembly, EMBO J 17, 918–926. 10. Muona, M., Aranko, A. S., and Iwai, H. (2008) Segmental isotopic labelling of a multidomain protein by protein ligation by protein trans-splicing, Chembiochem 9, 2958–2961. 11. Kurpiers, T., and Mootz, H. D. (2007) Regioselective cysteine bioconjugation by appending a labeled cystein tag to a protein by using protein splicing in trans, Angew Chem Int Ed Engl 46, 5234–5237. 12. Kurpiers, T., and Mootz, H. D. (2008) Sitespecific chemical modification of proteins with a prelabelled cysteine tag using the artificially split Mxe GyrA intein, Chembiochem 9, 2317–2325.

13. Wu, H., Xu, M. Q., and Liu, X. Q. (1998) Protein trans-splicing and functional miniinteins of a cyanobacterial dnaB intein, Biochim Biophys Acta 1387, 422–432. 14. Brenzel, S., Kurpiers, T., and Mootz, H. D. (2006) Engineering artificially split inteins for applications in protein chemistry: biochemical characterization of the split Ssp DnaB intein and comparison to the split Sce VMA intein, Biochemistry 45, 1571–1578. 15. Telenti, A., Southworth, M., Alcaide, F., Daugelat, S., Jacobs, W. R., Jr., and Perler, F. B. (1997) The Mycobacterium xenopi GyrA protein splicing element: characterization of a minimal intein, J Bacteriol 179, 6378–6382. 16. Brenzel, S., Cebi, M., Reiss, P., Koert, U., and Mootz, H. D. (2009) Expanding the scope of protein trans-splicing to fragment ligation of an integral membrane protein: towards modulation of porin-based ion channels by chemical modification, Chembiochem 10, 983–986. 17. Saves, I., Ozanne, V., Dietrich, J., and Masson, J. M. (2000) Inteins of Thermococcus fumicolans DNA polymerase are endonucleases with distinct enzymatic behaviors, J Biol Chem 275, 2335–2341. 18. Appleby, J. H., Zhou, K., Volkmann, G., and Liu, X. Q. (2009) Novel split intein for transsplicing synthetic peptide onto C terminus of protein, J Biol Chem 284, 6194–6199.

Part II Nucleic Acid Conjugates

Chapter 10 Polyethylenimine Bioconjugates for Imaging and DNA Delivery In Vivo Andrea Masotti and Francesco Pampaloni Abstract Polyamine polymers are among the commonest polymers used in biomedicine. Among polyamine ­polymers, polyethylenimine (PEI) may be used as an efficient delivery vehicle for nucleic acids (DNA, RNA, etc.) or employed as a versatile imaging probe in vivo. In this chapter, the preparation of various PEI bioconjugates will be fully explained and discussed. Key words: Polyamines, Polyethylenimine, Bioconjugate polymers, DNA delivery, Imaging

1. Introduction Polyethylenimine (PEI) is a versatile organic polymer widely employed in several biomedical applications (1), as well as being an efficient gene delivery vector (Fig. 1) (2, 3). PEIs, in both the linear and branched forms, have been used for the delivery of oligonucleotides (4), DNA, small RNA, and siRNA (5). The delivery efficiency generally varies accordingly to the type of polymer structure (linear or branched), molecular weight and degree of chemical substitution. By functionalizing PEI with alkyl chains (Fig. 2) or with an appropriate dye (Fig. 3), several interesting compounds, have recently been obtained (6, 7). For example, hydrophobic PEI derivatives give vesicular structures, called polycationic liposomes, that strongly interact with DNA. These compounds show a slow rate of release of DNA in vitro. PEI derivatized with a fluorescent dye (i.e., a near-infrared dye) allows one to follow the in vivo distribution of DNA in animal models. A PEI–dye conjugate containing the near-infrared emission fluorescent dye IR-820, an indocyanine derivative, has been obtained (IR820–PEI)

Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_10, © Springer Science+Business Media, LLC 2011

145

146

Masotti and Pampaloni

Fig. 1. The structure of branched polyethylenimine (PEI).

Fig. 2. Schematic of the functionalization of branched PEI with hydrophobic chains for the production of polycationic liposomes.

Polyethylenimine Bioconjugates for Imaging and DNA Delivery In Vivo

147

Fig. 3. Structure of the near-infrared (NIR) dye IR820–PEI conjugate.

(Fig. 3); this conjugate appears to be particularly promising for monitoring DNA delivery in vivo. The synthesis of substituted hydrophobic compounds, in conjunction with tagging with fluorescent dyes, allows one to obtain multifunctional delivery vectors with tunable properties. Specifically, the vector’s functional properties can be varied by choosing different hydrophobic grafting molecules, by adjusting the percentage of substitution, and by employing appropriate fluorophores. Tuning the properties of delivery vectors is valuable for many biomedical applications; the preparation and use of these derivatives will be explained and discussed in this chapter.

2. Materials 2.1. Reagents

1. Branched polyethylenimine (PEI) (25 kDa) (Sigma-Aldrich) (see Note 2). 2. Dichloromethane (DCM). 3. Dimethylformamide (DMF). 4. Triethylamine (TEA). 5. Ethyl acetate (AcOEt). 6. Methanol (MeOH). 7. Dimethyl sulfoxide (DMSO). 8. Chloroform (CHCl3).

148

Masotti and Pampaloni

9. Lauryl bromide (97%). 10. Myristyl bromide (97%). 11. Cetyl bromide (97%). 12. Lauroyl chloride (97%). 13. Myristoyl chloride (98%). 14. Palmitoyl chloride (98%). 15. Indocyanine dye IR-820 (Sigma-Aldrich). 16. 1-Ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC). 17. Poly-l-aspartic acid (PAA) (15–50 kDa) (Sigma-Aldrich). 18. Hoechst 33258. 19. Tris-Borate-EDTA (TBE) Buffer (1×): 90  mM Tris-borate, 2 mM ethylenediaminetetraacetic acid (EDTA), pH 8.3. 20. NIH-3T3 fibroblast cell line (ATCC). 21. Cell tissue culture plates (24-well and 96-well). 22. Dulbecco’s Modified Eagle’s Medium (DMEM) with 10% (v/v) fetal calf serum. 23. Animal models (e.g., nude mice). 24. CellTiter-Blue™ kit (Promega). 25. Plasmid DNA encoding enhanced green fluorescent protein (pEGFP) (Clontech). 26. Plasmid DNA encoding cytomegalovirus beta-galactosidase (pCMV b-Gal) (Clontech). 27. b-Galactosidase enzyme assay kit. 2.2. Equipment

1. Ultrafiltration apparatus. 2. Regenerated cellulose membranes (10 kDa MWCO). 3. Cellulose dialysis tubing (10 kDa MWCO). 4. Rotary evaporator (rotavap). 5. Lyophilizer. 6. Gel electrophoresis apparatus equipped with an image acquisition system. 7. Nuclear magnetic resonance (NMR) spectrometer. 8. Dynamic light scattering (DLS) instrument. 9. Potentiometric titration apparatus. 10. Laminar flow hood for cell culture. 11. Controlled-environment incubator (37°C) for culturing mammalian cells. 12. Fluorescence microplate reader. 13. Fluorescence microscope.

Polyethylenimine Bioconjugates for Imaging and DNA Delivery In Vivo

149

14. Transmission electron microscope (TEM) and ­Formvar-coated TEM grids. 15. Electrospray ionization (ESI) mass spectrometer (MS). 16. Optical imaging system equipped with near-infrared (NIR) fluorescence detector.

3. Methods The synthesis procedures to obtain various PEI derivatives consist of several steps that we will briefly illustrate in the following sections. We will separately describe the preparation of hydrophobic derivatives (Subheading  3.2) and a fluorescent derivative (Subheadings 3.2 and 3.4). Additional information may be found in the cited publications (see References section) 3.1. Polyethylenimine Solubilization

Branched high-molecular weight polyethylenimine (PEI) is a viscous polymer that is generally highly soluble in water, methanol, ethanol, chloroform, and other nonpolar solvents. The major drawback is the slow dissolution kinetics – branched PEI dissolves quite slowly in water. A convenient procedure to dissolve PEI is the following: 1. Use a spatula or a little spoon to collect a small amount of PEI directly from the product bottle. When ordering PEI, verify that the bottle has a wide-mouth opening. This will simplify the overall collection procedure. 2. Place the viscous polymer material into a small beaker (pretared on an analytical balance), taking care to distribute it onto one side of the beaker. 3. Weight the amount of PEI added to the beaker using an analytical balance. 4. Submerge the PEI with water (or other solvents) and place a magnetic stirrer into the beaker. 5. Stir gently until complete dissolution is achieved. Initially, stirring may become impaired by the high viscosity of the PEI. 6. Wait until the PEI material is no longer visible on the beaker’s sidewall. This indicates that the PEI has become fully dissolved.

3.2. Synthesis of PEI Derivatives by Grafting Reactions

In order to obtain PEI derivatives, the first step is to calculate the overall number of moles of the polymer to be reacted and defining the desired percentage level of substitution. For a given number of moles of grafting agent, an exact amount (moles) of PEI should be employed. However, when working with PEI ­polymers,

150

Masotti and Pampaloni

a precise molar ratio cannot be achieved, since the polymer itself does not have a single, unique molecular weight. Generally the molecular weight of a polymer such as PEI typically follows a Gaussian distribution. For example, branched PEI from SigmaAldrich (No. 408727) has a MW of ~25 kDa by light-scattering analysis or an average MW ~10 kDa by gel permeation chromatography (GPC). To calculate the molarity of a PEI solution, the common practice is to consider –(CH2–CH2–NH)– as the monomer unit (8) with a MW 43.068 Da. From this assumption, the number of monomeric units present in PEI 25  kDa is: ~25,000/43.068 = ~580. Since the monomer unit contains one nitrogen atom, it follows that PEI 25 kDa contains ~580 amino groups. As the calculated ratio of primary:secondary:tertiary amino groups in the polymer is 25:50:25, this implies that in the case of PEI 25 kDa the distribution of primary, secondary, and tertiary amino groups is, respectively, 145:290:145. This number is useful for calculating the desired percentage of substitution. For example, in order to obtain a PEI grafted at a 10% substitution level, ~58 groups have to be functionalized. As an example, we report here a detailed protocol for the functionalization of PEI with lauryl chains at a substitution percentage of 3%. The same protocol may be employed for the synthesis of other hydrophobic derivatives of PEI (at different percentage levels of substitution) by properly adjusting the stoichiometry of the starting reagents. For additional examples, see ref. (6). 1. Prepare a solution of lauryl bromide by adding 1-­bromododecane (0.178 g, 0.72 mmol) in DCM (50 mL) over a period of 3 h to a solution of PEI (1 g, 0.04 mmol) in DCM (100 mL). 2. Stir the resulting mixture at room temperature (RT) for 1 day. 3. Concentrate the solution by bringing down the solvent to a volume of ca. 15–20  mL under reduced pressure using a rotavap. The obtained product has a viscous aspect and a yellowish color. 4. The product may be purified by dialysis or by ultrafiltration (the second option is preferred). Dialyze the product against an ethanol/water (1:1) solution (4 L, with five changes of the solution), or perform ultrafiltration using 1  L of the same ethanol/water mixture with an ultrafiltration apparatus. 5. Dialyze the product again with pure water for 1 day, or perform ultrafiltration using water (3  L) with an ultrafiltration apparatus. 6. Transfer the dense, sticky oil to a round-bottom flask and proceed to lyophilize using standard procedures. Place the flask at −20°C until the product is completely frozen, then

Polyethylenimine Bioconjugates for Imaging and DNA Delivery In Vivo

151

attach the flask to a lyophilization apparatus. Lyophilize the product until all the water is removed. The final product should be a semicrystalline material. 3.3. Preparation of a Linker-Modified IR-820 Dye Intermediate [ 1]

For the preparation of fluorescent or near-infrared PEI conjugates, the same procedure used for polymer grafting is applied. However, a linker molecule between PEI and the dye must be employed as a spacer (Fig. 4). In fact, PEI is a branched polymer with amino groups localized both on the external surface and in the inner part. The coupling of the dye to the inner amino groups may result in decreased fluorescence. We have also observed that the direct conjugation of PEI with dyes and molecules may decrease the photostability of the dye itself, leading to partial degradation (data not shown). Thus, the attachment of a spacer molecule to the dye is recommended in order to avoid detrimental dye–PEI interactions. One of the most interesting dyes that we have conjugated to PEI is the near-infrared dye IR-820 (7). The IR-820 dye has a heptamethine moiety with heterocyclic nitrogen atoms bearing two

Fig. 4. Schematic of the synthesis of aminohexanoic acid-linked IR-820 ®e intermediate 1 and the NIR dye–polymer conjugate IR820–PEI.

152

Masotti and Pampaloni

alkyl-sulfonate groups. These pendant arms improve ­photostability and also provide a sphere of solvation in water that prevents dye aggregation. IR-820 dye also has a reactive chloride group. Linking an aminohexanoic acid molecule to IR-820 dye was successfully carried out to obtain a near-infrared PEI derivative (compound 1). In principle, any other molecules could be employed for this purpose. Aminohexanoic acid, having both one amino and one carboxylic group, may be used as a heterobifunctional linking molecule. In this case, we exploited the free amino group for dye functionalization to afford a versatile dye intermediate. The synthesis of compound 1 is performed as follows: 1. Place the IR-820 dye (300 mg, 0.283 mmol) and an excess of 6-aminohexanoic acid (126 mg, 1.42 mmol) inside a roundbottom flask. 2. Dissolve the mixture with 10 mL of anhydrous DMF under a nitrogen atmosphere. 3. Add an excess of TEA (198 mL, 1.42 mmol) to the mixture. 4. Using a stirring plate, incubate the flask at 85°C for 3 h. The color of the solution turns from green to blue during this time. 5. The product is recovered by solvent evaporation under reduced pressure, followed by flash chromatography purification on a silica gel column (AcOEt/MeOH from 70/30 to 0/100, v/v). 6. Characterize the linker-modified IR-820 dye molecule 1 by infrared (IR) spectroscopy, nuclear magnetic resonance (NMR) spectroscopy, and mass spectrometry (MS) (see Note 3). 3.4. Conjugation of PEI with Linker-Modified IR-820 [ 1]

The conjugation of PEI with the linker-modified dye molecule 1 follows a slightly different method compared to the alkyl chain grafting procedure (Fig. 4). In fact, since the reaction involves the formation of an amide bond, a peptide-like synthesis reaction should be performed. To this aim, a carboxylic group activator must be employed. We have found that the use of 1-ethyl-3-(3dimethylaminopropyl)carbodiimide (EDC) in the hydrochloride form is well suited to achieve this aim. This carbodiimide is water soluble and is typically employed in the pH 4.0–6.0 range. To obtain the polyethylenimine–dye molecule bioconjugate, the following steps are performed: 1. Dissolve PEI (84  mg, 0.0037  mmol) in anhydrous DMF (5 mL) and stir to the complete dissolution of the polymer. 2. Dissolve the linker-modified IR-8320 dye molecule 1 (19 mg, 0.020  mmol) in a round-bottom flask with 5  mL of DMF under a nitrogen atmosphere at room temperature.

Polyethylenimine Bioconjugates for Imaging and DNA Delivery In Vivo

153

3. Add a catalytic amount of EDC to the PEI solution and ­transfer the mixture to the linker-dye molecule 1 solution. 4. Stir the reaction at room temperature for 12 h. 5. Ultrafilter the reaction mixture on a cellulose membrane (10 kDa MWCO) against methanol. The IR820–PEI conjugate product is used as an aqueous solution after evaporation of the methanol. 6. Characterize the IR820–PEI conjugate by IR spectroscopy and NMR (see Note 4). 7. Store the IR820–PEI conjugate at −20°C protected from light (see Notes 5 and 6). 3.5. Ultrafiltration and Lyophilization of PEI Derivatives

These two procedures are used to completely purify the synthesized PEI derivative compounds (obtained from Subheading 3.2) and allow an easy recovery of the final product in a semicrystalline form. Some of the reported derivatives are reduced to fine powders that can facilitate weighing. A schematic representation of an ultrafiltration apparatus is shown in Fig. 5. 1. Thoroughly mix the PEI or PEI derivative compound with 200 mL of deionized water. 2. Place the resulting mixture into the ultrafiltration cell (Fig. 5). 3. Apply a nitrogen flux (4 atm.) to the top of the apparatus. 4. Collect the permeate solution into a clean beaker or a flask.

Fig. 5. Ultrafiltration apparatus used to purify the PEI derivative compounds after synthesis.

154

Masotti and Pampaloni

5. Collect the retentate solution containing the ultrafiltered polymer product. 6. Lyophilize the retentate solution to obtain a semicrystalline solid product. 3.6. 1H NMR Spectroscopy of PEI Derivatives

To assess the purity of the synthesized PEI derivative compounds (obtained from Subheading 3.2), NMR spectroscopy is the technique of choice. 1H and 13C NMR spectra were recorded on a 300-MHz NMR instrument using default acquisition parameters and standard deuterated solvents (water, chloroform, and reference materials). Several indications are reported to simplify the overall data acquisition procedure. 1. Dissolve a small amount of the derivatized polymer sample in D2O at a pH 14. 2. Place the solution into an NMR sample tube. 3. Acquire a 1H NMR spectrum of the sample according to your instrument set up. Integration of the proton magnetic resonance (1H NMR) spectrum of the product should indicate ~3 mol % of lauryl groups (C12H25; 1–1.6  ppm) per residue mol of ethylenimine unit (C2H4NH; 2.2–3.2 ppm) in the polymer. The substituted polymer may therefore be represented by the stoichiometric formula (C2H4NH)m(C12H25)0.03m, m = 580. The 1H and 13C NMR spectra of the product should be very close to the following example. 1H NMR (D2O): d (ppm) 0.92 (b, CH3), 1.30 (b, –(CH2)8–), 2.90 (b, –CH2CH2N–). 13 C NMR (D2O): d (ppm) 14.4, 23.1, 29.9, 30.2, 32.3, 37.8, 39.5, 44.0, 45.9, 47.7, 49.3, 51.4, 52.0, 53.1, 54.2, 69.2, 75.8.

3.7. Potentiometric Titration and Data Analysis of PEI Derivatives

PEI and PEI derivative polymer samples (obtained from Subheading 3.2) at a concentration of 45 mM (monomer aqueous solution) were titrated in a jacketed cell at 25°C at 0.1  M ionic strength (KCl) using a potentiometric apparatus equipped with a burette for automatic titration to control the addition of acid or base. The concentration of the HCl and NaOH solutions was standardized following reported procedures (9). 1. Place 10–20 mg of the PEI derivative into the potentiometric cell and dissolve it with 25 ml of KCl 0.1 M. 2. After dissolution, begin titrating the sample with HCl since the starting pH of the solution is alkaline. 3. Register and plot the pH values during the titration process for determination of the acid-base profile. As a representative example, in Fig. 6 we show the titration of a PEI derivative obtained by grafting the polymer with lauryl chains at different percentages of substitution.

Polyethylenimine Bioconjugates for Imaging and DNA Delivery In Vivo

155

Fig.  6. Representative titration data for PEI derivatives obtained by grafting PEI with dodecane chains at different percentage levels of substitution.

3.8. Agarose Gel Electrophoresis of DNA Polyplexes and Polyanion Exchange Reaction Assessment

Agarose gel electrophoresis was performed with a common gel electrophoresis apparatus equipped with a power supply. A digital camera and a digital recorder were used in all the described experiments to record the gel images. When preparing polymer/DNA complexes (polyplexes), the N/P ratio should be defined: The N/P ratio indicates the ionic balance between the PEI amino groups (N) and the DNA phosphate groups (P) in their complexes. The ratio is calculated based on considering that 1 mg DNA corresponds to 3 nmol of phosphate, and that 1 ml of PEI (or PEI derivative) at a concentration of 10 mM (monomers) corresponds to 10 nmol of amine nitrogen.

3.8.1. Agarose Gel Electrophoresis of Polymer/DNA Polyplexes

Gel shift assays of PEI and PEI derivatives complexed with DNA (polyplexes) are performed as follows: 1. Prepare a stock solution (10 mg/ml) of the intercalating dye Hoechst 33258 by dissolving the pure solid in a mixture of DMSO/H2O (2:1, v/v). Add 10 ml of dye stock solution to each 50 ml of 0.8% (w/v) agarose gel solution (in 1× TBE) used to cast the electrophoresis gel (see Note 1). 2. Prepare the PEI or PEI derivative solutions (1 mg/mL). 3. Add 1 mg plasmid DNA (e.g., pCMV b-Gal) to each mL of polymer solution. 4. Incubate the polymer/DNA mixture at 37°C for 5 min.

156

Masotti and Pampaloni

Fig. 7. Gel retardation assays (agarose 0.8%, 1× TBE, Hoechst 33258 staining) of ­polymer/ DNA polyplexes. (a) Analysis of polyplexes containing DNA and PEI or PEI derivatives with different percentage levels of substitution. All polyplex complexes were formed at a N/P ratio = 15.5. Samples: Reference DNA (without polymer) (Lane 1); PEI/DNA (Lane 2 ); Polymer/DNA complexes with PEI grafted with cetyl chains at 3% (Lane 3 ), 6% (Lane 4 ), 13% (Lane 5 ), and 30% (Lane 6 ) substitution. (b) Analysis of PEI grafted with myristoyl chains (3% substitution) complexed with DNA at N/P ratio = 15.5 after the addition of poly-aspartic acid (PAA). Samples: DNA molecular weight standards (ladder) (Lane 1); reference DNA (without polymer) (Lane 2 ); Polymer/DNA complexes after PAA addition: 10 min (Lane 3 ), 1 h (Lane 4 ), 2 h (lane 5 ), 3 h (lane 6 ), 18 h (lane 7 ) and 24 h (lane 6 ).

5. Add gel loading buffer to the polymer/DNA mixture. Load the sample onto the agarose gel (prepared in step 1) and run the gel for 45 min at 100 V (Fig. 7a). 3.8.2. Polyanion Exchange Reaction Assessment

In order to check the DNA complexing ability of PEI and the PEI derivatives, poly-l-aspartic acid (PAA) is used to competitively displace DNA from the polyplexes. Agarose gel electrophoresis is performed following the steps below: 1. Prepare the DNA/PEI or DNA/PEI derivative complexes as described above in Subheading 3.8.1. DNA is generally added to the PEI or PEI derivative solution (and not the other way around), as this facilitates the formation of DNA–polymer complexes. 2. Add 2 mg of poly-l-aspartic acid (PAA) for each microgram of DNA in the DNA/PEI or DNA/PEI derivative complexes. Incubate the mixtures at 37°C for 5 min.

Polyethylenimine Bioconjugates for Imaging and DNA Delivery In Vivo

157

3. Add running buffer to each sample and load the gel with solutions collected at 10 min, 1 h, 2 h, 3 h, 18 h, and 24 h after the addition of PAA (Fig. 7b). 3.9. Dynamic LightScattering Analysis of Hydrophobic PEI Derivatives and Their DNA Polyplexes

Dynamic light-scattering (DLS) experiments were performed with a classical setup equipped with a correlator operating with logarithmic sample time spacing. The light source was a 10-mW He–Ne laser (wavelength 632.8 nm) and the data were acquired at a scattering angle of 90°. The size and size distribution of the diffusing particles were found using the standard data analysis program CONTIN, based on fitting a continuous distribution of exponential decay times. The light-scattering data were used to assess whether the PEI derivatives or their DNA complexes (polyplexes) are able to form vesicular structures (polycationic liposomes) due to the presence of hydrophobic chains within the polymeric scaffold. 1. Weight 4–5 mg of a hydrophobic derivative of PEI (e.g., containing lauryl, myristyl, and cetyl chains) in a round-bottom flask. 2. Dissolve the product in a mixture of CHCl3/MeOH (3:1, v/v) (10 ml). 3. Evaporate the solvent with a rotavap until a thin film is obtained on the flask wall. 4. Add distilled water and vortex until the film is completely swelled and detached from the flask wall. 5. Transfer the sample into a light-scattering cuvette and start the data acquisition. As a representative example, Fig. 8a shows the dynamic lightscattering measurements for PEI derivatives containing lauryl, myristyl, and cetyl chains. The light-scattering profile of ­polyplexes obtained by grafting PEI with lauryl chains (3% of substitution) at different N/P ratios: (a) N/P = 15.5, (b) N/P = 11.6, (c) N/P = 7.75, (d) N/P = 3.87 is also reported (Fig. 8b). From these measurements we concluded that functionalizing PEI with hydrophobic chains imparts the polymer with the ability to form vesicular structures resembling liposomes. For this reason, these structures are also called polycationic liposomes.

3.10. Freeze-Fracture Microscopy Analysis

Hydrophobic PEI derivatives giving vesicular structures (as observed by light scattering) may also be examined by freeze-fracture microscopy. Conveniently, the same samples prepared for analysis by light scattering may also be used for freeze-fracture analysis. 1. Soak the PEI derivative sample with 30% (v/v) glycerol, and freeze them in partially solidified Freon 22. 2. Fracture the material in a freeze-fracture device (−105°C, 10−6 mmHg).

158

Masotti and Pampaloni

Fig. 8. Dynamic light-scattering measurements of (a) hydrophobic PEI derivatives (lauryl (a ), myristyl (b ) and cetyl (c )) at 3% substitution and (b) PEI derivatives obtained by grafting PEI with dodecane chains at 3% substitution under different N/P ratios: N/P = 15.5 (a), N/P = 11.6 (b), N/P = 7.75 (c), and N/P = 3.87 (d ).

Polyethylenimine Bioconjugates for Imaging and DNA Delivery In Vivo

159

Fig. 9. Freeze fracture images of a PEI derivative obtained by grafting PEI with tetradecane chains at 6% substitution outlining the polycation assembly into vesicles.

3. Replicate the sample by vacuum deposition of platinum/­ carbon using an electron beam gun. 4. Extensively wash the replica with distilled water and transfer it to a Formvar-coated TEM grid. 5. Examine the replica using a transmission electron microscope. As a representative example, Fig.  9 shows a freeze-fracture image of a PEI derivative obtained by grafting PEI with ­tetradecane chains at 6% substitution. The outlines of features resulting from the assembly of the PEI derivative into polycationic vesicles are visible. 3.11. Cytotoxicity, Transfection Assays, and Fluorescence Microscopy Analysis

Internalization assays were performed using mouse NIH-3T3 fibroblast cells to evaluate the cytotoxicity of PEI and hydrophobic PEI derivative compounds. The transfection assays were performed using PEI and PEI derivatives complexed with two reporter plasmids: The first plasmid encodes cytomegalovirus beta-galactosidase (pCMV b-Gal), and the second plasmid encodes enhanced green fluorescent protein (pEGFP). Both plasmids were obtained after bacterial amplification and purification from culture medium following standard protocols. Dulbecco’s Modified Eagle’s Medium (DMEM) supplemented with 10% (v/v) fetal calf serum was used in all cytotoxicity and transfection assays.

160

Masotti and Pampaloni

Fig.  10. NIH-3T3 cell viability assay results for PEI and a hydrophobic PEI derivative obtained by grafting PEI with lauryl chains at 3% substitution. 3.11.1. Cytotoxicity Assessment of PEI and PEI Derivatives

1. Plate  1 × 105 cells in a 96-well cell culture plate. Allow the cells to grow to 80% confluence. 2. Add gene delivery vehicle test compounds (hydrophobic PEI derivatives) or vehicle controls (unmodified PEI) (suspended in 100 ml medium) to each culture well. 3. Incubate the cells with the test compounds and controls in a mammalian cell culture incubator for 3 h at 37°C. 4. Remove the cell culture plates from the 37°C incubator and add CellTiter-Blue™ reagent (20  ml/well) to a blank well, wells containing treated cells, and wells containing untreated control cells. 5. Shake the cells with the CellTiter-Blue™ reagent for 10 s. 6. Incubate the cells using standard cell culture conditions for 1–4 h. 7. Shake the plate for 10  s and record the fluorescence (EX l = 560 nm, EM l = 590 nm). 8. Follow the manufacturer’s instructions for the CellTiter-Blue™ kit to calculate the results of the cell viability test (Fig. 10).

3.11.2. Transfection Assays with Polymer/DNA Polyplexes

1. Plate  2 × 105  cells in a 24-well cell culture plate. Allow the cells to grow to 80% confluence. 2. Prepare 500 mL of a solution containing 1 mg of pEGFP and 1 mg of pCMV b-Gal in medium to serve as a control for the evaluation of transfection efficiency. 3. In a separate tube, prepare 500 mL of a solution containing 1 mg of pEGFP and 1 mg of pCMV b-Gal complexed with

Polyethylenimine Bioconjugates for Imaging and DNA Delivery In Vivo

161

2  mg of PEI or equivalent amounts of the different ­hydrophobic PEI derivatives. 4. Incubate the solutions prepared in steps 2 and 3 above at RT for 10 min. 5. Add the DNA and polymer/DNA polyplex solutions (500 mL) to the desired wells in the culture plate containing cells. 6. Incubate the cells for 3 h and then replace the medium with fresh medium. 7. Examine the GFP fluorescence under a fluorescence microscope to confirm that transfection has occurred. 8. Assess b-galactosidase activity in the whole cell extracts by following the instructions provided with the employed b-galactosidase enzyme assay kit (Fig. 11a).

Fig. 11. (a) b-gal expression efficiency of NIH-3T3 cells transfected with PEI and a PEI derivative obtained by grafting PEI with hexadecane chains. The b-gal activity is measured in OD units 24 h after transfection. (b) Fluorescence microscopy of NIH-3T3 cells transfected with PEI and a PEI derivative obtained by grafting PEI with hexadecane chains.

162

Masotti and Pampaloni

3.11.3. Fluorescence Microscopy

1. Perform the procedures described in steps 1–5 in Subheading 3.11.2 above. 2. Harvest the cells 24 h after transfection. 3. Fix the cells and stain with Hoechst 33258 (1 mg/ml in PBS, pH 7.4). Hoechst 33258 fluorescently stains double-stranded DNA in cells. 4. Examine the cells under a fluorescence microscope using 40× magnification (Fig. 11b).

3.12. NIR Optical Imaging

Near-infrared (NIR)-emitting dyes represent an intriguing avenue for extracting biological information from living subjects since they can be monitored with safe, noninvasive optical imaging and contrast techniques (10, 11). Optical imaging represents a rapidly expanding field, with direct applications in ­pharmacology, molecular and cellular biology, as well as diagnostics. For the ­evaluation of DNA delivery using PEI derivatives, we used a standard NIR optical imaging device to image nude mice injected with a solution of fluorescent conjugates (Fig. 12). 1. Inject nude mice (200–250 g each) via the caudal vein with a 2 mg/ml solution of fluorescent conjugates at a 0.015 mg/g dosage. 2. After injection, place the anesthetized animal under the optical imaging instrument. 3. Acquire a photograph of the animal before irradiation to serves as the “blank” image. The default imaging parameters used in this work were as follows: Exposure time = 0.2  s; ­binning = 4; f = 8; field of view (FOV) = 12.8 × 12.8 cm.

Fig. 12. Optical images of nude mice subjected to tail vein injection of various polymer and polymer/DNA complexes. The images show the distribution of the IR820–PEI vector alone, the IR820–PEI/DNA complex and PEI/IR820–DNA. A threedimensional (3D) reconstruction of a mouse body indicating the organs of interest (brain, lungs, and liver) is also shown.

Polyethylenimine Bioconjugates for Imaging and DNA Delivery In Vivo

163

4. Acquire NIR optical images for 3 h at a rate of 1 image/min. The default imaging parameters used in this work were as follows: Excitation filter ICG (710–760  nm); emission filter ICG (810–875 nm); exposure time = 1 s; binning = 16, f = 2, FOV = 12.8 × 12.8 cm.

4. Notes 1. Replacing ethidium bromide (EtBr) with Hoechst 33258 in the agarose gel. For the agarose gel electrophoresis procedure, a slight modification has been introduced with respect to the classical setup. Generally, EtBr is added to the agarose gel during gel casting. As a result, the nucleic acids incorporate the dye during electrophoresis and become fluorescent under UV illumination. However, some authors have reported that when DNA/EtBr interacts with PEI, an almost complete quenching of the dye fluorescence is observed (12). This quenching effect is reduced by ~50% when a minor groove-dye such as Hoechst 33258 is employed. For this reason, we incorporated Hoechst 33258 instead of EtBr during gel casting. 2. Commercially available high-molecular weight PEI polymers may contain different amounts of water. Linear PEI (25 kDa) is available as a pure crystalline solid, while the branched form is a viscous liquid containing traces of water. PEI with a molecular weight of 500 kDa is generally available as a 50:50 mixture with water. Several other different formulations for PEI may be found commercially; therefore, carefully check the amount of water present in the compound and calculate the exact concentration of the polymer accordingly. 3. The linker-modified IR-820 dye molecule 1 shows the following characteristics: IR (film): 1,718, 1,740 cm−1. 1H NMR (DMSO-d6): d (ppm) 8.10 (m, 3H), 7.89 (m, 4H), 7.66 (d, J = 12  Hz, 2H), 7.50 (m, 4H), 7.31 (m, 2H), 5.67 (d, J = 12 Hz, 2H), 3.96 (m, 4H), 3.05 (m, 4H), 2.53 (m, 4H), 1.96 (m, 4H), 1.86 (s, 12H), 1.76 (m, 6H), 1.68 (s, 6H), 1.44 (m, 4H). 13C NMR (DMSO-d6): d (ppm) 177.2, 166.6, 165.0, 162.5 (2C), 140.9, 130.2, 129.9 (4C), 128.4 (4C), 127.3 (4C), 123.3 (4C), 121.7 (4C), 110.8 (2C), 93.7, 51.0 (2C), 50.2, 48.8 (2C), 38.2, 37.1, 36.0, 35.9, 30.9, 30.7, 27.7 (4C), 27.5, 26.4, 25.7, 25.2 (2C), 22.6 (2C). The elemental analysis calculated for C52H62N3NaO8S2 is: C, 66,15; H, 6,62; N, 4,45; Na, 2,43; O, 13,56; S, 6,79%. Found: C, 66, 23; H, 6,87; N, 4,47; S, 6,77%. The ESI-MS spectra resulted in a peak with m/z = 921.56 relative to [M-Na]−. 4. The conjugation reaction between PEI and the linker-modified dye molecule 1 yields the derivative IR820–PEI, which shows

164

Masotti and Pampaloni

the following characteristics: IR (CHCl3): 1,670  cm−1. The 1 H NMR spectrum (CDCl3) of IR820–PEI was poorly resolved, but the signal integration indicated the presence of 1 mol % of IR820 (CH2; 0.9 ppm) per residue mol of ethylenimine unit (C2H4NH; 2.2–3.2 ppm) in the polymer. The substituted polymer may be represented by the stoichiometric formula (C2H4NH)m(C52H61N3NaO7S2)0.01m, m = 580. 5. When storing the dye–polymer conjugate, dye photobleaching processes should be prevented. Hence, we recommend wrapping aluminum foil all around the container (bottle, flask, vials, etc.) used to store the product and keeping the product at −20°C. 6. A brief comment regarding collateral NIR dye oxidation: Bouteiller et al. recently described the synthesis and the photophysical properties of two novel near-infrared (NIR) ­cyanine dyes, NIR5.5-2 and NIR7.0-2. These two dyes are water soluble and are potential substitutes for the commercially available Cy5.5 and Cy7.0 dyes (13). It has been found that meso-chlorine derivatives of stabilized Cy7.0-like dyes conjugated to amino acids are not stable in Milli-Q grade water; in particular, the cleavage of the amino acid residue and subsequent conversion of the dye into an alpha-hydroxy ketone (or related hydroperoxide) has been observed. The decomposition mechanism is similar to the one that has been recently suggested by Toutchkine et al. (14) to explain the photodecomposition of merocyanine dyes.

Acknowledgments The Ministero della Salute and MIUR (Ministero dell’Istruzione, dell’Università e della Ricerca) are gratefully acknowledged for providing financial support for this work. FP also thanks Ernst H.K. Stelzer for support and many interesting discussions. References 1. Vicennati, P., Giuliano, A., Ortaggi, G., and Masotti, A. (2008) Polyethylenimine in medicinal chemistry. Curr. Med. Chem. 15, 2826–2839. 2. Eliyahu, H., Barenholz, Y., and Domb, A. J. (2005) Polymers for DNA delivery. Molecules 10, 34–64. 3. Park, T. G., Jeong, J. H., and Kim, S. W. (2006) Current status of polymeric gene delivery systems. Adv. Drug Deliv. Rev. 58, 467–486.

4. Lemkine, G. F., and Demeneix, B. A. (2001) Polyethylenimines for in  vivo gene delivery. Curr. Opin. Mol. Ther. 3, 178–182. 5. Jere, D., Jiang, H. L., Arote, R., Kim, Y. K., Choi, Y. J., Cho, M. H., Akaike, T., and Cho, C. S. (2009) Degradable polyethylenimines as DNA and small interfering RNA carriers. Expert Opin. Drug Deliv. 6, 827–834. 6. Masotti, A., Moretti, F., Mancini, F., Russo, G., Di Lauro, N., Checchia, P., Marianecci, C.,

Polyethylenimine Bioconjugates for Imaging and DNA Delivery In Vivo Carafa, M., Santucci, E., and Ortaggi, G. (2007) Physicochemical and biological study of selected hydrophobic polyethyleniminebased polycationic liposomes and their complexes with DNA. Bioorg. Med. Chem. 15, 1504–1515. 7. Masotti, A., Vicennati, P., Boschi, F., Calderan, L., Sbarbati, A., and Ortaggi, G. (2008) A novel near-infrared indocyanine dye-polyethylenimine conjugate allows DNA delivery imaging in  vivo. Bioconjug. Chem. 19, 983–987. 8. Boussif, O., Lezoualc’h, F., Zanta, M. A., Mergny, M. D., Scherman, D., Demeneix, B., and Behr, J. P. (1995) A versatile vector for gene and oligonucleotide transfer into cells in culture and in  vivo: Polyethylenimine. Proc. Natl. Acad. Sci. USA. 92, 7297–7301. 9. Gans, P., and O’Sullivan, B. (2000) GLEE, a new computer program for glass electrode calibration. Talanta 51, 33–37. 10. North Atlantic Treaty Organization. Scientific Affairs Division. (1998) Near-infrared dyes for high technology applications. In: Daehne, S., Resch-Genger, U., and Wolfbeis, O.S. (Eds.)

165

Proceedings of the NATO Advanced Research Workshop on Syntheses, Optical Properties and Applications of Near-Infrared (NIR) Dyes in High Technology Fields, Trest, Czech Republic, September 24–27, 1997. Kluwer, Dordrecht, London. 11. Mishra, A., Behera, R. K., Behera, P. K., Mishra, B. K., and Behera, G. B. (2000) Cyanines during the 1990s: A review. Chem. Rev. 100, 1973–2012. 12. Wiethoff, C. M., Gill, M. L., Koe, G. S., Koe, J. G., and Middaugh, C. R. (2003) A fluorescence study of the structure and accessibility of plasmid DNA condensed with cationic gene delivery vehicles. J. Pharm. Sci. 92, 1272–1285. 13. Bouteiller, C., Clave, G., Bernardin, A., Chipon, B., Massonneau, M., Renard, P. Y., and Romieu, A. (2007) Novel water-soluble near-infrared cyanine dyes: Synthesis, spectral properties, and use in the preparation of internally quenched fluorescent probes. Bioconjug. Chem. 18, 1303–1317. 14. Toutchkine, A., Nguyen, D. V., and Hahn, K. M. (2007) Merocyanine dyes with improved photostability. Org. Lett. 9, 2775–2777.

Chapter 11 Synthesis of a Glycomimetic Oligonucleotide Conjugate by 1,3-Dipolar Cycloaddition Gwladys Pourceau, Albert Meyer, Jean-Jacques Vasseur, and François Morvan Abstract A glycomimetic oligonucleotide conjugate bearing four galactose residues on a mannose core is ­synthesized using oligonucleotide solid-phase synthesis and Cu(I)-catalyzed azide-alkyne 1,3-dipolar cycloaddition (CuAAC, or “click” chemistry). To achieve this purpose, new building blocks (including the solid support and phosphoramidites) are synthesized and used on a DNA synthesizer to generate a tetraalkyne oligonucleotide, which is then conjugated with a galactose azide derivative by click chemistry to afford the desired 3¢-tetragalactosyl-mannose oligonucleotide conjugate. The procedures described in this chapter provide a general approach for the synthesis of novel glycoconjugates that can be immobilized to a DNA chip via DNA-directed immobilization to study, for example, their multivalent interactions with lectins in cellular targeting/uptake, etc. Key words: Solid-phase automated oligonucleotide synthesis, 1,3-Dipolar cycloaddition, Click chemistry, Glycoconjugate, Glycocluster, Glycomimetic, Tetraalkyne oligonucleotide, Galactose azide

1. Introduction Interactions between carbohydrates and sugar-binding proteins – known as lectins – are involved in many biological processes and play a leading role in cellular recognition events such as those involved in viral and bacterial adhesion (1–3). For example, Pseudomonas aeruginosa (PA) is an opportunistic pathogen that is largely involved in the morbidity and mortality of cystic fibrosis and immunocompromised patients. This clinically important bacterium expresses two soluble lectins, PA-IL and PA-IIL, which specifically recognize d-galactose and l-fucose, respectively (4). Treatment

Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_11, © Springer Science+Business Media, LLC 2011

167

168

Pourceau et al.

with the corresponding saccharide derivatives has been shown to be potent against acute pneumonia in mice models (5); however, in order to find highly effective inhibitors of PA, the synthesis of novel glycomimetics of galactose is required (6). Furthermore, it is necessary to be able to characterize the interactions between glycomimetics and lectins to understand the relevant biological processes and develop new therapeutic strategies. To achieve this purpose, researchers need to adequately address several different challenges. First, the great structural diversity of monosaccharides and their large number of branching sites generally make polysaccharide synthesis procedures difficult to perform. Moreover, the interactions between lectins and carbohydrates are relatively weak and require multipresentation of the relevant residues to be significant (7); hence, many types of glycoclusters have been designed and reported in literature. To overcome the synthesis difficulties and to take advantage of the so-called cluster effect, a number of simple carbohydrate derivatives have been conjugated with various scaffolds such as linear (8) or cyclic (9) peptides, calixarenes (10, 11), oligosaccharides (12), pentaerythritol (13), and cyclodextrins (14); however, very few examples of glycoclusters based on DNA oligonucleotide chemistry currently exist (15–20). Thanks to the phosphoramidite method developed by Beaucage and Caruthers in 1981 (21), the use of solid-phase oligonucleotide synthesis and the development of automated DNA synthesizers has grown considerably. We thus decided to take advantage of these efficient synthetic tools and combined them with methods based on Cu(I)-mediated azide-alkyne 1,3-dipolar cycloaddition (CuAAC) (22, 23) to synthesize new glycocluster conjugates using a non-nucleosidic alkyne phosphoramidite and an azide solid support. In addition, the glycocluster conjugates can be immobilized to a DNA chip via DNA-directed immobilization (DDI) (24, 25) to study, for example, their multivalent interactions with lectins in cellular targeting/uptake, etc. In this chapter, we describe a strategy to synthesize a novel 3¢-tetragalactose oligonucleotide conjugate using a tetraalkyne mannose core that is conjugated with a galactose azide derivative by CuAAC. To achieve the synthesis of the conjugate, it is necessary to synthesize previously the following: (1) a solid support exhibiting an azide group and a protected hydroxyl group (26); (2) 1-O-propargyl-a-d-mannopyranose (27, 28) for its introduction onto the solid support by CuAAC, thus providing a solidsupported mannose core with four free hydroxyls; (3) a pentynyl phosphoramidite (26) to introduce alkyne groups into the mannose core; and (4) a galactose azide derivative (29) for its introduction four times by a second round of CuAAC reactions onto the tetraalkyne mannose core present at the 3¢-end of the oligonucleotide synthesized by solid phase. Using these specific building blocks, we describe herein the synthesis of a 3¢-tetragalactose oligonucleotide conjugate; however, by modifying the relevant

Synthesis of a Glycomimetic Oligonucleotide Conjugate by 1,3-Dipolar Cycloaddition

169

procedures accordingly, it is obviously possible to introduce other types of saccharides using the corresponding azide derivatives. Since the CuAAC reaction is highly efficient, similar yield levels can be expected in such cases regardless of the particular type of saccharide azide derivative that is utilized. Moreover, by following similar procedures, non-saccharidic azide derivatives could also be used to afford the synthesis of new oligonucleotide conjugates exhibiting other types of biomolecules such as peptides, amines/polyamines, lipids, and biotinylated ligands (30).

2. Materials 2.1. Equipments

1. Rotary evaporator. 2. Silica gel (0.04–0.06 mm particle size). 3. Vacuum desiccator with phosphorous pentoxide (P2O5). 4. UV–Visible spectrophotometer. 5. Microwave synthesizer with 0.2–0.5-mL vials, caps, and crimper. 6. ABI 394 DNA synthesizer with synthesis columns (Applied Biosystems). 7. SpeedVac® vacuum concentrator system. 8. Matrix-assisted laser desorption/ionization-time of flight (MALDI-TOF) mass spectrometer. 9. Reversed-phase high-performance liquid chromatography (HPLC) system. 10. Reversed-phase C18 (5 mm) column (150 × 4.6-mm; for analytical HPLC). 11. Reversed-phase C18 (15  mm) column (300 × 7.8-mm; for preparative-scale HPLC purification). 12. NAP™-10 disposable columns prepacked with Sephadex™ G-25 (GE Healthcare).

2.2. Synthesis of Carbohydrate Derivatives 9 and 19

1. Peracetylated d-mannose. 2. Peracetylated d-galactose. 3. Propargyl alcohol. 4. Boron trifluoride etherate (BF3∙OEt2). 5. 1,4-Cyclohexanedimethanol. 6. p-Toluenesulfonyl chloride. 7. Sodium azide (NaN3). 8. Sodium iodide (NaI).

170

Pourceau et al.

2.3. Synthesis of Modified Solid Support 6

1. 1,1,1-Tris(hydroxymethyl)ethane. 2. 4,4¢-Dimethoxytrityl chloride. 3. Sodium hydride (NaH). 4. Sodium iodide (NaI). 5. 1,6-Dibromohexane. 6. Tetramethylguanidinium azide (TMGA). 7. 4-Dimethylaminopyridine (DMAP). 8. N-(3-Dimethylaminopropyl)-N ¢-ethylcarbodiimide (DEC). 9. Hydroquinone-O,O ¢-diacetic acid (“Q-linker”). 10. Long-chain alkylamine controlled-pore glass (LCAA-CPG, 500 Å, 135 mmol amino group/g loading). 11. Acetic anhydride. 12. Cap A solution: Acetic anhydride/dry pyridine/THF, 10/10/80 (v/v/v). 13. Cap B solution: 10% (v/v) N-Methylimidazole in dry THF.

2.4. Synthesis of O-Pent-4-ynyl-O ¢(2-cyanoethyl)N,N ¢-diisopropylphos­ phoramidite 12 2.5. DNA Synthesis

Pent-4-yn-1-ol 1. Diisopropylammonium tetrazolide (DIAT). 2. 2-Cyanoethyl-(N,N,N¢,N¢-tetraisopropyl)phosphorodia­ midite. 1. Standard 2-cyanoethyl deoxyribonucleoside phosphoramidites can be obtained from commercial suppliers. 2. Activator solution: 0.3 M 5-Benzylthio-1H-tetrazole (BMT) in dry acetonitrile. 3. Oxidation solution: 0.1  M Iodine in 70/10/20 (v/v/v) ­tetrahydrofuran (THF)/pyridine/water. 4. Cap A solution: Acetic anhydride/dry pyridine/THF, 10/10/80 (v/v/v). 5. Cap B solution: 10% (v/v) N-Methylimidazole in dry THF. 6. Deblocking solution: 3% (v/v) Dichloroacetic acid in dry CH2Cl2.

2.6. Click Reactions

1. Copper sulfate (CuSO4). 2. Sodium ascorbate.

2.7. Dry Solvents

1. Pyridine. 2. Methylene chloride (CH2Cl2). 3. Dry and extra dry acetonitrile (CH3CN). 4. Triethylamine (Et3N). 5. Tetrahydrofuran.

Synthesis of a Glycomimetic Oligonucleotide Conjugate by 1,3-Dipolar Cycloaddition

2.8. “For Synthesis” Quality Solvents

171

1. Pyridine. 2. Methylene chloride (CH2Cl2). 3. Acetonitrile (CH3CN). 4. Triethylamine (Et3N). 5. Cyclohexane. 6. Ethyl acetate (EtOAc). 7. Methanol (MeOH). 8. Toluene. 9. Acetone. 10. Petroleum ether (Ep). 11. Concentrated aqueous ammonia solution (NH4OH). 12. Dimethylformamide (DMF).

2.9. Buffers and Other Solutions

1. Saturated aqueous sodium bicarbonate solution. 2. Saturated aqueous sodium chloride solution (brine). 3. 0.05 M Triethylammonium acetate, pH 7 (TEAAc). 4. 0.1 M p-Toluenesulfonic acid in acetonitrile (CH3CN). 5. Sulfuric acid (H2SO4) stain solution: Ethanol (EtOH)/H2SO4 (90:10, v/v). 6. Vanillin stain solution: Add 5  g of vanillin and 25  mL of H2SO4 to 225 mL of MeOH/H2O (1:1, v/v).

3. Methods The first few sections described below outline the synthesis, purification, and characterization/quantification of the following building blocks: the mannose-core solid support 10 (Subheading 3.1); the O-pent-4-ynyl-O ¢-(2-cyanoethyl)-N,N-diisopropylphosphoramidite 12 (Subheading 3.2); the 3¢-(tetrapentynylphosphodiester mannosyl) oligonucleotide 14 (Subheading  3.3); and the 1-O-[4(azidomethyl)cyclohexyl-1-methyl]-2,3,4,6-tetra-O-acetylb-d-galactopyranose 19 (Sub­heading 3.4.1).The conjugation of the galactosyl azido derivative 19 by CuAAC with the 3¢-(tetrapentynylmannose)-oligonucleotide intermediate 14 to generate the final 3¢-tetragalactose oligonucleotide conjugate product 20 (containing a mannose core) is described in Subheading 3.4.2. 3.1. Preparation of Mannose-Core Solid Support 10

The synthesis of the mannose-core solid support 10 is described below in Subheadings  3.1.1–3.1.3. The synthesis procedure requires first the preparation of the solid support azide 6 (26) and the 1-O-propargyl-b-d-mannopyranose 9 (27, 28), which are then reacted together by microwave-assisted CuAAC to provide 10 (Fig. 1).

172

Pourceau et al. HO

HO

DMTrCl

HO

Pyr

OH

Br (CH2)6 Br

DMTr O

1

OH

Br (CH2)6 O

THF NaH

DMTr O

2

3 NH2 N

O HO2C N3 (CH2)6

O

N

N3 (CH2)6

O

DMTr

O

CH3CN

N3

DEC pyr DMAP

O

OH 4

O

O

O

5

CO2H

O

O

DMTr

OH

OAc

CO2H

AcO O

AcO AcO NH2

CPG

N3 (CH2)6

OAc 7

DEC pyr DMAP

BF3. Et2O

OH

CH2Cl2

O

DMTr

O

OAc

O O

O

O

HN

AcO O

AcO AcO

8

O

6

O NH4OH MeOH

OH HO

CuSO4 Na Ascorbate H2O/MeOH MW

HO HO 9

OH HO HO HO

O

N

N

N (CH2)6

O

O

DMTr 10

O

O O

O

O

HN O

Fig. 1. Synthesis of the solid support mannose core 10.

O O

Synthesis of a Glycomimetic Oligonucleotide Conjugate by 1,3-Dipolar Cycloaddition 3.1.1. Preparation of Azido Solid Support 6 3.1.1.1. Dimethoxytritylation of Tris(hydroxymethyl) ethane 1

173

1. In a 500-mL round-bottom flask, coevaporate 3.6  g (30 mmol, 2 eq.) of 1,1,1-tris(hydroxymethyl)ethane 1 three times, each with 50  mL of dry pyridine. Dissolve the final evaporated residue in 250 mL of dry pyridine. 2. While stirring, add 5.1 g (15 mmol, 1 eq.) of 4,4¢-dimethoxytrityl chloride under an argon (Ar) atmosphere. 3. Stir the reaction mixture overnight. Monitor by thin layer chromatography (TLC) (see Notes 1–4) and stain with sulfuric acid solution. (TLC: CH2Cl2/MeOH/Et3N, 99:1:1 v/v/v; Rf DMTrCl = 0.7; Rf2 = 0.15.) 4. Stop the reaction with 10  mL of MeOH to hydrolyze any remaining 4,4¢-dimethoxytrityl chloride which did not react. Concentrate the mixture to dryness on a rotary evaporator under reduced pressure and dissolve the residue in 100 mL of CH2Cl2. 5. Pour the reaction mixture into a 250-mL separatory funnel and wash the organic layer twice with 100 mL of saturated aqueous sodium bicarbonate solution. 6. Collect the organic layer, dry it by adding 10 g of anhydrous sodium sulfate, and then filter through a cotton-plugged funnel. Concentrate on a rotary evaporator under reduced pressure and coevaporate with toluene to remove pyridine residues; a yellow oil is obtained. 7. Dissolve the crude product in a minimal volume of CH2Cl2 and apply the solution to a 5-cm diameter chromatography column containing 120 g of silica gel equilibrated firstly with 250 mL of CH2Cl2/Et3N, (99:1, v/v) and then with 250 mL of CH2Cl2 (see Note 5). 8. Gradually increase the concentration of acetone (from 1 to 50%, v/v) in CH2Cl2. 9. Monitor the collected fractions by TLC and combine those containing the pure compound. 10. Evaporate to dryness on a vacuum evaporator to obtain a foam. 11. Check the purity of the synthesized compound by 1H, NMR and MS.

C

13

2-[(4,4¢-Dimethoxytrityl)oxymethyl]-2-methylpropane-1,3diol 2: Yield 53%. TLC (CH2Cl2/MeOH/Et3N, 99:1:1 v/v/v): Rf = 0.15. 1H NMR (CDCl3, 400 MHz): 0.87 (3H, s), 2.42 (2H, br s), 3.16 (2H, s), 3.69 (2H, d, J = 25.2  Hz), 3.64 (2H, d, J = 25.2 Hz), 3.82 (6H, s), and 6.79–7.49 (13H, m). 13 C NMR (CDCl3, 100 MHz): 17.4, 41.1, 55.2, 67.2, 68.2, 86.3, 113.3, 126.9, 128.0, 130.0, 135.8, 144.7, and 158.5. HRMS FAB (positive mode, nitrobenzylic alcohol) m/z: calcd for C26H31O5 [M + H]+ 422.2093; found 422.2098.

174

Pourceau et al.

3.1.1.2. Alkylation of 1-O-[(4,4¢Dimethoxytrityl) oxymethyl]-2methylpropane-1,3-diol 2

1. In a 50-mL round-bottom flask, coevaporate 844  mg (2  mmol, 1  eq.) of 2 three times, each with 10  mL of dry CH3CN. Dissolve the final evaporated residue in 25 mL of dry THF. 2. While stirring at 0°C, add 240 mg (6 mmol, 3 eq.) of 60% (w/w) sodium hydride (NaH) in mineral oil, 30  mg (0.2  mmol, 0.1  eq.) of sodium iodide (NaI), and 1.54  mL (10 mmol, 5 eq.) of 1,6-dibromohexane. 3. Stir the mixture at room temperature overnight. Monitor by TLC (see Notes 1, 3, 4) and stain with sulfuric acid solution. (TLC: cyclohexane/acetone, 1:1 v/v; Rf2 = 0.5; Rf3 = 0.6.) 4. Dilute the reaction mixture with 20 mL of CH2Cl2 and then carefully stop the reaction with 3 mL of water to hydrolyze the excess NaH. 5. Pour the reaction mixture into a 250-mL separatory funnel and wash the organic layer twice with 100 mL of saturated aqueous sodium bicarbonate solution. 6. Collect the organic layer, dry it by adding 10 g of anhydrous sodium sulfate, and filter through a cotton-plugged funnel. Concentrate on a rotary evaporator under reduced pressure to obtain a yellow oil. 7. Dissolve the crude product in a minimal volume of 99:1 (v/v) cyclohexane/Et3N and apply the solution to a 5-cm diameter chromatography column containing 75 g of silica gel equilibrated with the same solvent. 8. Gradually increase the concentration of ethyl acetate (from 0 to 50%, v/v) in cyclohexane, still containing 1% (v/v) of Et3N (see Note 5). 9. Monitor the collected fractions by TLC and combine those containing the pure compound. 10. Evaporate to dryness on a vacuum evaporator to obtain a foam. 11. Check the purity of the synthesized compound by 1H, NMR and MS.

C

13

1-O-(4,4¢-Dimethoxytrityl)-2-(6-bromohexyloxymethyl)-2methylpropane-1,3-diol 3: Yield 69%. TLC (cyclohexane/ acetone, 1:1 v/v): Rf = 0.6. 1H NMR (CD3CN, 300 MHz): 0.89 (3H, s), 1.34–1.54 (6H, m), 1.78–1.88(2H, m), 2.67 (1H, t, J = 5.5 Hz), 2.98 (2H, s), 3.36–3.49 (8H, m), 3.79 (6H, s), and 6.87–7.48 (13 H, m). 13C NMR (CD3CN, 75 MHz): 16.8, 24.7, 26.6, 27.3, 28.9, 32.0, 32.2, 33.8, 33.9, 54.5, 65.0, 66.1, 70.7, 73.8, 85.0, 112.5, 126.3, 127.4, 127.7, 129.7, 136.0, 136.1, 145.2, and 158.2. HRMS ESI (positive mode) m/z: calcd for C32H41O5BrNa [M + Na]+ 607.2035; found 607.2083.

Synthesis of a Glycomimetic Oligonucleotide Conjugate by 1,3-Dipolar Cycloaddition 3.1.1.3. Azidation

175

1. In a 25-mL round-bottom flask equipped with a condenser and a CaCl2 guard, dissolve 1.17  g (2  mmol, 1  eq.) of 1-O-(4,4¢-dimethoxytrityl)-2-(6-bromohexyloxymethyl)-2methylpropane-1,3-diol 3 in 5 mL of dry CH3CN. 2. While stirring, add 949.2 mg (6 mmol, 3 eq.) of TMGA. 3. Stir the mixture 3 h at 50°C. Monitor by TLC (see Notes 1, 3, 4, 6) and stain with sulfuric acid solution. (TLC: cyclohexane/EtOAc, 85:15 v/v; Rf3 = 0.46; Rf4 = 0.41.) 4. Stop the reaction with 50  mL of CH2Cl2 and wash with 70 mL of brine. 5. Pour the reaction mixture into a 250-mL separatory funnel and extract the aqueous layer twice with 50 mL of CH2Cl2. 6. Collect the organic layers, dry them by adding 10 g of anhydrous sodium sulfate, and filter through a cotton-plugged funnel. Concentrate the mixture on a rotary evaporator under reduced pressure. 7. Dissolve the crude product in a minimal volume of 99:1 (v/v) cyclohexane/Et3N and apply the solution to a 5-cm diameter chromatography column containing 30 g of silica gel equilibrated with the same solvent. 8. Gradually increase the concentration of EtOAc (from 0 to 30%, v/v) in cyclohexane, still containing 1% (v/v) of Et3N (see Note 5). 9. Monitor the collected fractions by TLC and combine those containing the pure compound. 10. Evaporate to dryness on a vacuum evaporator to obtain a foam. 11. Check the purity of the synthesized compound by 1H, NMR and MS.

C

13

1-O-(4,4¢-Dimethoxytrityl)-2-(6-azidohexyloxymethyl)-2methylpropane-1,3-diol 4: Yield 52%. TLC (cyclohexane/EtOAc, 85:15 v/v): Rf = 0.41. 1H NMR (CD3CN, 300 MHz): 0.90 (3H, s), 1.29–1.59 (8H, m), 2.67 (1H, br s), 2.99 (2H, s), 3.28 (2H, t, J = 6.9  Hz), 3.36–3.44 (6H, m), 3.77 (6H, s), and 6.86–7.49 (13H, m). 13C (CD3CN, 75 MHz): 16.8, 25.1, 25.9, 28.1, 28.9, 40.6, 50.8, 54.5, 65.0, 66.1, 70.7, 73.8, 85.1, 112.6, 126.3, 127.4, 127.8, 129.8, 136.0, 145.3, and 158.2. HRMS ESI (positive mode) m/z: calcd for C32H41N3O5Na [M + Na]+ 570.2944; found 570.2955. 3.1.1.4. Anchoring of 1-O-(4,4¢Dimethoxytrityl)-2(6-azidohexyloxymethyl)2-methylpropane-1,3-diol 4 on LCAA-CPG Beads

1. In a 25-mL round-bottom flask, coevaporate 344  mg (0.63  mmol, 1  eq.) of 1-O-(4,4¢-dimethoxytrityl)-2-(6azidohexyloxymethyl)-2-methylpropane-1,3-diol 4 and 15.4 mg (0.13 mmol, 0.2 eq.) of 4-DMAP three times, each time with 10  mL of dry pyridine. Dissolve the final evaporated residue in 5 mL of dry pyridine.

176

Pourceau et al.

2. While stirring, add 170.5  mg (0.75  mmol, 1.2  eq.) of hydroquinone-O,O ¢-diacetic acid (“Q-linker”) (31) and 120.2 mg (0.63 mmol, 1 eq.) of 1-(3-dimethylaminopropyl) 3-ethylcarbodiimide (DEC) under an Ar atmosphere. 3. Stir the mixture at room temperature overnight. Monitor by TLC (see Notes 1–4) and stain with sulfuric acid solution. (TLC: CH2Cl2/MeOH, 99:1 v/v: Rf4 = 0.36; Rf5 = 0.) 4. Pour the reaction mixture into a 500-mL separatory funnel with 150 mL of water and 150 mL of CH2Cl2. Extract the aqueous layer twice with CH2Cl2, each time with 150 mL. 5. Collect the organic layers, dry them by adding 10 g of anhydrous sodium sulfate, and filter through a cotton-plugged funnel. Concentrate on a rotary evaporator under reduced pressure. 6. Dissolve the crude product in a minimal volume of 98.5:1:0.5 (v/v/v) CH2Cl2/pyridine/MeOH and apply the solution to a 5-cm diameter chromatography containing 15 g of silica gel equilibrated with the same solvent. 7. Gradually increase the concentration of MeOH (from 0.5 to 10%, v/v) in CH2Cl2, still containing 1% (v/v) of pyridine (see Note 5). 8. Monitor the collected fractions by TLC and combine those containing the pure compound. 9. Evaporate to dryness on a vacuum evaporator. (Yield: 23%). 10. In a sealed tube, coevaporate 500 mg of LCAA-CPG (70 mmol of NH2, 1 eq.) and 10 mg (70 mmol, 1 eq.) of DMAP three times with dry pyridine, each time with 10 mL. 11. Add 120 mg of 5 (140 mmol, 2 eq.) dissolved in 4 mL of dry pyridine, and then add 120 mg (0.7 mmol, 10 eq.) of DEC and 80 mL of dry Et3N. Gently shake the mixture overnight at room temperature. 12. Filter off the CPG beads on a frit and wash them with 10 mL of CH2Cl2. 13. Dry the recovered CPG beads in a vacuum desiccator over P2O5 for 30 min at room temperature (see Note 7). 14. In a sealed tube, add successively the CPG beads, 5  mL of commercial Cap A solution and 5 mL of commercial Cap B solution. Gently shake the mixture for 1  h at room temperature. 15. Filter off the CPG beads on a frit and wash them three times, each time with 10  mL of CH2Cl2. Dry the recovered CPG beads in a vacuum desiccator over P2O5 for 30 min at room temperature, affording 6.

Synthesis of a Glycomimetic Oligonucleotide Conjugate by 1,3-Dipolar Cycloaddition

177

16. Accurately weigh out ~10  mg of 6. Transfer the weighed amount of 6 into a 10-mL volumetric flask and fill up to 10 mL with 0.1 M p-toluenesulfonic acid solution in CH3CN. Transfer 1 mL of the resulting solution to a clean 10-mL volumetric flask and add 9  mL of the same p-toluenesulfonic solution. 17. Measure the absorbance of the mixture at 498 nm (A498) and calculate the loading in micromoles per gram of CPG (mmol/g CPG) according to the following formula: Loading(mmol / g CPG ) = ( A 498 ´ 10 ´ 10 ´ 106 ) / (70, 000 ´ m ) where m is the mass of 6, “70,000” is the extinction coefficient of the cation DMTr, and the two “10” values correspond to the total volume of the solution (mL) and the dilution factor. In our hands, the loading of the solid support 6 was calculated to be ~61 mmol/g CPG (see Note 8). 3.1.2. Preparation of 1-O-Propargyl-a-dmannopyranose 9 3.1.2.1. Glycosylation of 2,3,4,6-O-Tetraacetyl-a-dmannopyranose 7

1. In a 50-mL round-bottom flask, coevaporate 1.9 g (4.8 mmol, 1 eq.) of peracetylated d-mannose 7 three times with 10 mL dry CH3CN. Dissolve the final evaporated residue in 20 mL of dry CH2Cl2 with 420 mg (7.5 mmol, 1.5 eq.) of propargyl alcohol. Cool the reaction mixture at 0°C. 2. While stirring at 0°C, add 3.14 mL (25 mmol, 5 eq.) of boron trifluoride etherate (BF3∙OEt2) under an Ar atmosphere. 3. Stir the mixture for 36 h at room temperature. Monitor by TLC (see Note 4) and stain with sulfuric acid solution. (TLC: Ep/EtOAc, 4:6 v/v; Rf7 = 0.48; Rf8 = 0.60.) 4. Dilute the reaction mixture with 10 mL of CH2Cl2 and then carefully stop the reaction with 30 mL of saturated aqueous sodium bicarbonate solution to hydrolyze any remaining BF3∙OEt2 that did not react. 5. Pour the reaction mixture into a 250-mL separatory funnel and extract the aqueous layer twice with 100 mL of CH2Cl2. 6. Collect the organic layers, dry them by adding 10 g of anhydrous sodium sulfate, and filter through a cotton-plugged funnel. Concentrate on a rotary evaporator under reduced pressure to obtain a yellow oil. 7. Dissolve the crude product in a minimal volume of 6:4 (v/v) cyclohexane/EtOAc and apply the solution to a 2.5-cm diameter chromatography column containing 40  g of silica gel equilibrated with the same solvent. 8. Gradually increase the concentration of EtOAc (from 40 to 60%, v/v) in cyclohexane. 9. Monitor the collected fractions by TLC and combine those containing the pure compound.

178

Pourceau et al.

10. Evaporate to dryness on a vacuum evaporator, dissolve the residue in a minimum of EtOAc, and then crystallize the white powder by adding petroleum ether. 11. Check the purity of the synthesized compound 8 (27) by 1H NMR. 1-O-Propargyl-2,3,4,6-tetraacetyl-a-d-mannopyranose 8: Yield 66%. TLC (Ep/EtOAc 4:6, v/v): Rf = 0.60. 1H NMR (CDCl3, 200 MHz): 1.65 (s), 2.03–2.21 (12H, 4 s), 2.51 (1H, t, J = 2.4 Hz), 4.07–4.38 (5H, m), and 5.07–5.41 (4H, m). 3.1.2.2. Hydrolysis of Acetyl Groups

1. In a 25-mL round-bottom flask, dissolve 503.6 mg, (1.3 mmol) of 1-O-propargyl-2,3,4,6-O-tetraacetyl-a-d-mannopyranose 8 in 14 mL of a mixture of NH4OH/MeOH (1:1, v/v). 2. Stir the mixture at 35°C for 2 h. Monitor by TLC (see Note 4) and stain with sulfuric acid solution. (TLC: Cyclohexane/ EtOAc 4:6 v/v; Rf 8 = 0.60; Rf  9 = 0). 3. Stop the reaction by evaporation of NH3, and concentrate on a rotary evaporator under reduced pressure. 4. Pour the reaction mixture into a 250-mL separatory funnel, wash the organic layer twice with 50 mL of water, and combine the aqueous layers. 5. Concentrate on a rotary evaporator under reduced pressure, coevaporate three times with 10  mL of CH3CN to remove water, and then dry the mixture in a vacuum desiccator over P2O5 overnight at room temperature. 1-O-Propargyl-a-d-mannopyranose 9: Yield 100%. HRMS ESI (positive mode): m/z calcd for C9H15O6 [M + H]+ 219.0869; found 219.0876.

3.1.3. Click Reaction Between Azido Solid Support 6 and 1-O-Propargyl-a-dmannopyranose 9

1. In a 2-mL microcentrifuge tube, prepare a 0.1-M stock solution of 1-O-propargyl-a-d-mannopyranose 6 by dissolving 21.8 mg in 1.0 mL of degassed water. 2. In separate 0.5-mL microcentrifuge tubes, prepare a 0.1-M solution of copper sulfate (CuSO4) (3.2  mg in 200  mL of degassed water) and a 0.5-M solution of sodium ascorbate (9.9 mg in 100 mL of degassed water). Mix 4 mL of the CuSO4 solution with 4  mL of the sodium ascorbate solution in a 0.5-mL microcentrifuge tube (see Note 9). 3. To a 0.2- to 0.5-mL microwave vial equipped with a micro stir bar and containing 16.4 mg (1 mmol, 1 eq.) of the azido solid support 6, add 50  mL of 1-O-propargyl-a-dmannopyranose solution (5 mmol, 5 eq.), 42 mL of degassed water, 100 mL of MeOH, and the mixture (8 mL) of sodium ascorbate (2 mmol, 2 eq.) and CuSO4 (0.4 mmol, 0.4 eq.) prepared in step 2.

Synthesis of a Glycomimetic Oligonucleotide Conjugate by 1,3-Dipolar Cycloaddition

179

4. Flush with Ar and seal the microwave vial with a cap using the crimper. Place the vial in the microwave synthesizer for 30 min at 70°C (see Note 10). 5. Transfer the beads and supernatant into a synthesis column using micropipettes (see Note 11). Connect a 2-mL syringe to the bottom of the opened column and remove the supernatant. Close the synthesis column, connect a second 2-mL syringe to the top, and wash the beads by pushing back and forth three times with 2 mL of water, three times with 2 mL of MeOH, and three times with 2 mL of CH3CN. Perform each wash for 1 min. Flush with dry nitrogen (N2) for 3 min. 6. Dry in a vacuum desiccator over P2O5 for 1 h at room temperature (see Note 12). 3.2. Preparation of O-Pent-4-ynylO ¢-(2-cyanoethyl)N,N-diisopropyl­ phosphoramidite 12

The pentynyl phosphoramidite (26) is prepared in a single procedure, starting from commercially available reagents as follows (Fig. 2): 1. Dry 420 mg (5 mmol, 1 eq.) of pent-4-yn-1-ol over molecular sieves (3 Å) overnight. 2. In a 50-mL round bottom flask, coevaporate 428  mg (2.5 mmol, 0.5 eq.) of DIAT three times with 10 mL of dry CH3CN. Dissolve the final evaporated residue in 15  mL of dry CH2Cl2. Add successively the pent-4-yn-1-ol (with molecular sieves) and 1.5 g (5 mmol, 1 eq.) of 2-cyanoethyl(N,N,N¢,N¢tetraisopropyl)-phosphorodiamidite. Equip the round-bottom flask with a CaCl2 guard. 3. Stir the mixture at room temperature for 4  h. Monitor by TLC (see Notes 1, 3, 4) and stain with vanillin solution. (TLC: Cyclohexane/CH2Cl2/Et3N, 5:4:1 v/v/v; Rf11 = 0.34; Rf12 = 0.65.) 4. Stop the reaction by adding 20 mL of CH2Cl2. OH 11

iPr2N

P

CH2Cl2

OCne

N N

iPr2N

N

N

iPr2NH2

O iPr2N

P

OCne

12

Fig. 2. Synthesis of O-pent-4-ynyl-O ¢-(2-cyanoethyl)-N,N-diisopropylphosphoramidite 12.

180

Pourceau et al.

5. Pour the reaction mixture into a 250-mL separatory funnel and wash the organic layer twice with 200 mL of brine. 6. Collect the organic layer, dry it by adding 10 g of anhydrous sodium sulfate, and filter through a cotton-plugged funnel. Concentrate on a rotary evaporator under reduced pressure to obtain a colorless oil. 7. Dissolve the crude phosphoramidite product in a minimal volume of 75:20:5 (v/v/v) cyclohexane/CH2Cl2/Et3N and apply the solution to a 2.5-cm diameter chromatography column containing 45 g of silica gel equilibrated with the same eluent (see Note 5). 8. Purify the crude product by isocratic chromatography using the same eluent. Analyze the collected fractions by TLC and combine those containing the pure compound. Evaporate to dryness on rotary evaporator to obtain a colorless oil. 9. Check the purity of the synthesized compound by 1H, 31 P NMR and MS.

C,

13

O-Pent-4-ynyl-O¢-(2-cyanoethyl)-N,Ndiisopropylphosphoramidite 12: Yield 80%. TLC (cyclohexane/ CH2Cl2/Et3N, 5:4:1  v/v/v): Rf = 0.65. 1H NMR (CDCl3, 200 MHz): 1.18–1.21 (12H, d, J = 6.4 Hz), 1.77–1.97 (3 H, m), 2.27–2.36 (2H, m), 2.62–2.69 (2H, m), and 3.55–3.88 (6H, m); 13C NMR (CDCl3, 75 MHz): 15.3, 20.4, 29.2, 42.1, 58.2, 58.4, 61.9, 68.9, 83.7, and 117.6; 31P NMR (CDCl3, 81 MHz): 148.8. HRMS ESI (positive mode) m/z: calcd for C14H26O2N2P [M + H]+ 285.1732; found 285.1732. 3.3. Synthesis of 3 ¢-(Tetrapentynyl­ phosphodiester mannosyl) Oligonucleotide 14 3.3.1. Elongation Cycle

This section describes the automated synthesis of an oligonucleotide bearing a tetrapentynylphosphodiester mannosyl at its 3¢end starting from the solid support 10, and using the pentynyl phosphoramidite 12 and commercially available nucleoside phosphoramidites on an ABI DNA synthesizer (Fig. 3). 1. Prepare a 0.3 M solution of BMT in anhydrous CH3CN (12 g in 200 mL). Place the bottle at position #9 in the DNA synthesizer using the change bottle procedure. Place the Cap A and Cap B solutions, the deblocking solution, and the oxidation solution at bottle position #11, 12, 14, and 15, respectively. 2. Prepare 0.075  M solutions of commercially available phosphoramidites in extra dry CH3CN and place the corresponding bottles on the DNA synthesizer. 3. Prepare a 0.2 M solution of phosphoramidite 12 in extra dry CH3CN and place it at position #5 using the change bottle procedure.

181

Synthesis of a Glycomimetic Oligonucleotide Conjugate by 1,3-Dipolar Cycloaddition 10 1) 12

2) I2 /H2O

BMT CH3CN

3) Ac2O

O O P OCne O O O O P OCne O O O P N CneO O O N N O P OCne O O

(CH2)6

O

DMTr

13

O

O O

O

O

1-SPOS

HN O

2-NH4OH

O O P O- O O P O O OO O O P N -O O O N N O P OO O

(CH2)6

O O-

ACA CCC AAT TCT O P

O

OH

O 14

Fig. 3. Synthesis of 3¢-(tetrapentynylphosphodiester mannosyl) oligonucleotide 14.

4. Program a modified elongation cycle for the ABI 394 DNA synthesizer as shown in Table 1 (see Note 13). 5. Enter the sequence of the desired oligonucleotide: 5¢-5T-3¢. In this program, 5 refers to the position of the modified phosphoramidite 12 and T refers to the solid support. 6. Pack a synthesis column with 1 mmol of the modified solid support 10 (16.4 mg). Place the column on the synthesizer. 7. Run the synthesis program corresponding to the reaction of 12 on 10 activated with BMT, then the oxidation of the resulting phosphite triester by means of the oxidation solution, and then the final capping step, affording 13.

182

Pourceau et al.

Table 1 Modified elongation cycle for the incorporation of pentynyl phosphoramidites using an ABI 394 DNA synthesizer Step

Reagents and solvents

Wash

CH3CN

10

Wash

CH2Cl2

35

Wash

CH3CN

20

Coupling

0.2 M amidite 12 + 0.3 M BMT in CH3CN

Wait

Time (s)

6 180

Wash

CH3CN

Coupling

0.2 M amidite 12 + 0.3 M BMT in CH3CN

Wait

15 6 180

Wash

CH3CN

15

Capping

Cap A and Cap B solution

10

Wait

15

Wash

CH3CN

Oxidation

Oxidation solution

Wait Wash

10 8 13

CH3CN

20

8. Program a standard elongation cycle for the ABI 394 DNA synthesizer as shown in Table  2. The standard elongation cycle is available through the synthesizer’s program library. 9. Enter the sequence of the desired oligonucleotide: 5¢-ACA CCC AAT TCT T-3¢. In this program, the first 3¢-T refers to the support 13. 10. Run the synthesis program with “Trityl off.” The elongation cycle corresponds to (1) a detritylation step; (2) coupling of the nucleoside phosphoramidite; (3) oxidation of the phosphite triester; and (4) capping (see Note 14). 3.3.2. Deprotection/ Cleavage and Analysis by HPLC and MALDITOF MS 3.3.2.1. Deprotection/ Cleavage

1. Transfer 8.2 mg (~0.5 mmol) of the solid-supported 3¢-tetrapentynyl-mannose oligonucleotide into a sealed HPLC vial. Add 3  mL of concentrated aqueous ammonia (NH4OH) solution and place the vial into a dry bath for 2 h at 65°C (see Note 15). 2. Recover the supernatant and evaporate the ammonia using a SpeedVac® vacuum concentrator system.

Synthesis of a Glycomimetic Oligonucleotide Conjugate by 1,3-Dipolar Cycloaddition

183

Table 2 Standard oligonucleotide elongation cycle for an ABI 394 DNA synthesizer Step

Reagents and solvents

Time (s)

Wash

CH3CN

10

Wash

CH2Cl2

35

Detritylation

Deblocking solution

43

Wash

CH3CN

20

Coupling

0.075 M commercial amidites + 0.3 M BMT in CH3CN

 6

Wait

20

Wash

CH3CN

15

Capping

Cap A and Cap B solution

10

Wait

15

Wash

CH3CN

10

Oxidation

Oxidation solution

 8

Wait Wash

3.3.2.2. Analysis by HPLC and MALDI-TOF MS

13 CH3CN

20

1. Dissolve the crude oligonucleotide 14 in 500 mL of pure water to determine its amount by UV spectrophotometry as follows: Withdraw 20 mL of the solution and add it to 1 mL of pure water. Read the absorbance at 260 nm using a UV–visible spectrophotometer (see Note 16). (Isolated amount: 178 nmol.) 2. Characterize 14 by MALDI-TOF mass spectrometry (matrix: 2¢,4¢,6¢-trihydroxyacetophenone, THAP) and by reversedphase HPLC (gradient: 8–32% (v/v) of CH3CN in 0.05 M TEAAc for 20 min). (HPLC RT = 10.26 min, lmax = 262 nm, e = 124,700 L/mol/cm, MALDI-TOF MS (negative mode) m/z calcd for C155H211N44O92P16 [M − H]−: 4658.202; found 4659.02.) 3. Freeze-dry the solution. The oligonucleotide 14 can be stored for several months in a freezer at −20°C under an Ar atmosphere.

3.4. Conjugation of Galactosyl Azide 19 to Tetraalkyne Oligonucleotide 14

This section describes the synthesis of the galactosyl azido derivative 19 (29) (Fig. 4), which is eventually conjugated in solution by CuAAC to the 3¢-(tetrapentynyl-mannose)-oligonucleotide 14 (Fig. 5) affording the final tetragalactosyl-mannose ­oligonucleotide conjugate product 20 after HPLC purification and ammonia treatment to remove the acetyl protecting groups (see Note 17).

184

Pourceau et al. HO

HO

Tos-Cl

H

DMAP Et3N

H OH

HO

NaN3 NaI

H

H

DMF

H

CH2Cl2

H

OTos

15

17

16

N3

AcO OAc

BF3. Et2O

O AcO 18

OAc

CH2Cl2

OAc

AcO OAc AcO

N3

H O

H

O

OAc

19

Fig.  4. Synthesis of 1-O-[4-(azidomethyl)cyclohexyl-1-methyl]-2,3,4,6-tetra-O-acetylb-d-galactopyranose 19.

14 19

CuSO4 Na Ascorbate H2O/MeOH

NH4OH

O HO

OH

O

N

N

N

O PO OO O O P O OP O O O N -O O O N N (CH2)6 O P OO O

H N N N

O

OH

O

H

H N N N

O

H

HO OH HO

HO

H

HO OH

H

H O

HO

O

N H

N

HO OH

O

OO ACA CCC AAT TCT O P O

HO OH

OH

O

OH

N 20

OH

Fig. 5. Synthesis of the tetragalactosyl-mannose oligonucleotide conjugate 20 by 1,3-dipolar cycloaddition.

Synthesis of a Glycomimetic Oligonucleotide Conjugate by 1,3-Dipolar Cycloaddition

185

3.4.1. Preparation of 1-O-[4-(Azidomethyl) cyclohexyl-1-methyl]2,3,4,6-tetra-O-acetyl-bd-galactopyranose 19

1. In a 50-mL round-bottom flask, coevaporate 1.4 g (10 mmol, 2  eq.) of 1,4-cyclohexanedimethanol 15 and 61  mg (0.5 mmol, 0.1 eq.) of DMAP three times, each with 25 mL of dry pyridine. Dissolve the final evaporated residue in 15 mL of dry CH2Cl2.

3.4.1.1. Prepare [4-(p-Toluenesulfonylmethyl) cyclohexyl]methanol 16

2. While stirring, add 950 mg (5 mmol, 1 eq.) of p-toluenesulfonyl chloride and 1.05 mL (7.5 mmol, 1.5 eq.) of dry Et3N under an Ar atmosphere. 3. Stir the mixture during 3 h. Monitor by TLC (see Notes 2–4) and stain with sulfuric acid solution. (TLC: CH2Cl2/MeOH, 95:1 v/v; Rf TosCl = 0.9; Rf15 = 0.2; Rf16 = 0.55.) 4. Stop the reaction with 50 mL of saturated aqueous sodium bicarbonate solution. 5. Pour the reaction mixture into a 250-mL separatory funnel and extract the aqueous layer twice with 100 mL of CH2Cl2. 6. Collect the organic layers, dry them by adding 10 g of anhydrous sodium sulfate, and filter through a cotton-plugged funnel. Concentrate on a rotary evaporator under reduced pressure. A pink oil is obtained. 7. Dissolve the crude in a minimal volume of CH2Cl2 and apply the solution to a 2.5-cm diameter chromatography column containing 35 g of silica gel equilibrated with CH2Cl2. 8. Gradually increase the concentration of MeOH (from 0 to 5%, v/v) in CH2Cl2. 9. Monitor the collected fractions by TLC and combine those containing the pure compound. 10. Evaporate to dryness on a vacuum evaporator to obtain a pink oil. 11. Check the purity of the synthesized compound by 1H, NMR and MS.

13

C

[4-(p-Toluenesulfonylmethyl)cyclohexyl]methanol 16: Yield 72%. TLC (CH2Cl2/MeOH 95:5  v/v): Rf = 0.55. 1H NMR (CDCl3, 300  MHz): 0.85–1.95 (11H, m), 2.47 (3H, s), 3.44–3.95 (2H, m), 3.83–3.95 (2H, m), 7.36 (2H, d, J = 8.1 Hz), and 7.80 (2H, d, J = 8.3 Hz). 13C NMR (CDCl3, 75 MHz): 22.0, 25.2, 25.3, 28.8, 35.0, 37.7, 38.1, 40.5, 66.3, 68.7, 73.5, 75.6, 128.3, 130.2, 133.4, and 145.0. HRMS ESI (positive mode) m/z: calcd for C15H23O4S1 [M + H]+ 299.1317; found 299.1305. 3.4.1.2. Prepare [4-(Azidomethyl)cyclohexyl] methanol 17

1. In a 100-mL round-bottom flask equipped with a condenser, dissolve 1 g (3.4 mmol, 1 eq.) of [4-(p-toluenesulfonylmethyl) cyclohexyl]methanol 16 in 25 mL of DMF. 2. While stirring, add 2.1 g (13.7 mmol, 4 eq.) of sodium iodide (NaI) and 890  mg (13.7  mmol, 4  eq.) of sodium azide (NaN3) at room temperature.

186

Pourceau et al.

3. Stir the mixture for 3 h at 75°C. Monitor by TLC (see Notes 3 and 4) and stain with vanillin solution. (TLC: Cyclohexane/ EtOAc 7:3, v/v; Rf16 = 0.1; Rf17 = 0.3.) 4. Remove the solvent under reduced pressure and dissolve the resulting residue in 100 mL of CH2Cl2. 5. Pour the reaction mixture into a 250-mL separatory funnel and wash the organic layer twice with 50  mL of saturated aqueous sodium bicarbonate solution and twice with 50 mL brine. 6. Collect the organic layer, dry it by adding 10 g of anhydrous sodium sulfate, and filter through a cotton-plugged funnel. Concentrate on a rotary evaporator under reduced pressure to obtain a yellow oil. 7. Dissolve the crude in a minimal volume of 93:7 (v/v) cyclohexane/EtOAc and apply the solution to a 2.5-cm diameter chromatography column containing 50 g silica gel equilibrated with the same solvent. 8. Gradually increase the concentration of EtOAc (from 7to 60%, v/v) in cyclohexane. 9. Monitor the collected fractions by TLC and combine those containing the pure compound. 10. Evaporate to dryness on a vacuum evaporator to obtain 17 as colorless oil. 11. Check the purity of the synthesized compound by 1H, 13C NMR, MS and Fourier transform infrared spectroscopy (FTIR) (see Note 18). [4-(Azidomethyl)cyclohexyl]methanol 17 Yield 79%. TLC (cyclohexane/EtOAc 7:3  v/v) Rf: 0.29. 1H NMR (CDCl3, 300  MHz): 0.85–1.95 (11H, m), 3.15–3.26 (2H, m), and 3.47–3.57 (2H, m). 13C NMR (CDCl3, 75  MHz): 25.1, 26.3, 28.8, 30.0, 35.5, 37.9, 38.3, 40.4, 55.4, 58.0, 66.0, and 68.5. HRMS ESI (positive) m/z: calcd for C8H16 N3O [M + H]+ 170.1293; found 170.1315. FTIR: 3343, 2922, 2856, 2524, 2096, 1451, 1378, 1345, 1262, 1132, 1097, 1036, 992, 955, 941, 885, 825, 655, 588, and 556. 3.4.1.3. Preparation of 1-O-[4-(Azidomethyl) cyclohexyl-1-methyl]2,3,4,6-tetra-O-acetyl-bd-galactopyranose 19

1. In a 50-mL round-bottom flask, coevaporate 414  mg (1.05 mmol, 1.05 eq.) of peracetylated d-galactose 18 three times with 10 mL dry acetonitrile. Dissolve 18 in 10 mL of dry CH2Cl2 with 171 mg (1 mmol, 1 eq.) of [4-(azidomethyl) cyclohexyl]methanol 17. Cool the reaction mixture to 0°C. 2. While stirring at 0°C, add 630 mL (5 mmol, 5 eq.) of boron trifluoride etherate (BF3∙OEt2) under an Ar atmosphere. 3. Stir the mixture 4 h at room temperature. Monitor by TLC (see Note 4) and stain with vanillin solution. (TLC:

Synthesis of a Glycomimetic Oligonucleotide Conjugate by 1,3-Dipolar Cycloaddition

187

Cyclohexane/EtOAc, 1:1  v/v; Rf18 = 0.27; Rf17 = 0.33; Rf19 = 0.40.) 4. Dilute the reaction mixture with 10 mL of CH2Cl2 and stop the reaction carefully with 30 mL of saturated aqueous sodium bicarbonate solution to hydrolyze the excess BF3∙OEt2. 5. Pour the reaction mixture into a 250-mL separatory funnel and extract the aqueous layer twice with 100 mL of CH2Cl2. 6. Collect the organic layers, dry them by adding 10 g of anhydrous sodium sulfate, and filter through a cotton-plugged funnel. Concentrate on a rotary evaporator under reduced pressure to obtain a yellow oil. 7. Dissolve the crude in a minimal volume of cyclohexane (see Note 19) and apply the solution to a 2.5-cm diameter chromatography column containing 25 g of silica gel equilibrated with cyclohexane. 8. Gradually increase the concentration of EtOAc (from 0 to 30%, v/v) in cyclohexane. 9. Monitor the collected fractions by TLC and combine those containing the pure compound. 10. Evaporate to dryness on a vacuum evaporator to obtain 19 as colorless oil. 11. Check the purity of the synthesized compound by 1H, NMR and MS.

C

13

1-O-[[4-(Azidomethyl)cyclohexyl]methyl]-2,3,4,6-tetra-Oacetyl-b-d-galactopyranose 19: Yield 21%. TLC (cyclohexane/ EtOAc, 1:1  v/v): Rf = 0.29. 1H NMR (CDCl3, 200  MHz): 0.85–2.08 (22H, m), 3.05–3.35 (3H, m), 3.65–3.86 (2H, m), 4.00–4.17 (2H, m), 4.34–4.39 (1H, m), 4.94 (1H, dd, J = 3.2 Hz, J = 10.5  Hz), 5.14 (1H, dd, J = 7.8  Hz, J = 10.5  Hz), and 5.32 (1H, d, J = 3.2  Hz). 13C NMR (CDCl3, 75  MHz): 19.9, 20.0, 20.3, 24.3, 24.4, 25.3, 25.5, 28.0, 28.2, 29.1, 29.2, 34.2, 34.8, 37.0, 37.4, 54.7, 57.1, 60.5, 66.3, 68.2, 69.9, 70.2, 72.3, 74.8, 100.9, 168.7, 169.5, 169.6, and 169.7. HRMS ESI (positive) m/z: calcd for C22H34N3O10 [M + H]+ 500.2244; found 500.2244. 3.4.2. Click Reaction of Galactosyl Azido Derivative 19 with 3¢-(Tetrapentynylmannose)-oligonucleotide 14 in Solution

1. In a 2-mL microcentrifuge tube, prepare a 1-mM solution of tetraalkyne mannose-oligonucleotide 14 (178  nmol in 178 mL of pure water). 2. In a 2-mL microcentrifuge tube, prepare a 0.1-M stock solution of 19 by dissolving 49.9 mg in 1 mL of MeOH. 3. In separate 0.5-mL microcentrifuge tubes, prepare a 0.04-M solution of copper sulfate (CuSO4) (6.4  mg in 1  mL of degassed water) and a 0.05-M solution of sodium ascorbate (9.9 mg in 1 mL of degassed water). Mix 5 mL of the CuSO4

188

Pourceau et al.

solution with 20  mL of the sodium ascorbate solution in a 0.5-mL microcentrifuge tube (see Note 9). 4. To a 0.2- to 0.5-mL microwave vial equipped with a micro stir bar, add 100  mL of the oligonucleotide solution 14 (100 nmol, 1 eq.), 113 mL of MeOH, 12 mL of 19 (1.2 mmol, 12 eq.), and the mixture (25 mL) of sodium ascorbate (1 mmol, 10 eq.) and CuSO4 (200 nmol, 2 eq.) prepared in step 3. 5. Flush with Ar and seal the microwave vial with a cap using the crimper. Place the vial in the microwave synthesizer for 45 min at 70°C. (See Note 20). 6. Desalt the resulting solution on a NAP™-10 column and then concentrate using a SpeedVac® vacuum concentrator system. 7. Dissolve the residue in 300 mL of pure water for characterization by MALDI-TOF mass spectrometry and by reversedphase HPLC (gradient: 5–60% (v/v) CH3CN in 0.05  M TEAAc for 20 min). (HPLC RT = 15.33 min, MALDI-TOF MS (negative mode) m/z: calcd for C243H343N56O132P16 [M − H]−: 6656.20; found 6653.95.) 3.4.3. Purification and Characterization of the Final Tetragalactosyl-Mannose Oligonucleotide Conjugate Product 20 3.4.3.1. Purification

3.4.3.2. Deprotection of Galactose Hydroxyls and Characterization

1. Purify the crude product obtained from Subheading 3.4.2 by preparative reversed-phase HPLC using a gradient of 32–44% (v/v) CH3CN in 0.05 M TEAAc for 15 min. 2. Pool the fractions containing the protected oligonucleotide glycoconjugate and evaporate to dryness. Coevaporate the residue further ten times, each with 1.5 mL of pure water to remove volatile buffer salts. (MALDI-TOF MS (negative mode) m/z: calcd for C243H343N56O132P16 [M − H]−: 6656.20; found 6656.74.) 1. In a 2-mL microcentrifuge tube, add successively the protected tetragalactose oligonucleotide conjugate obtained after HPLC purification and 1 mL of concentrated aqueous ammonia solution. Stir the reaction mixture at room temperature for 2 h. 2. Evaporate the ammonia using a SpeedVac® vacuum concentrator system. 3. Dissolve the residue in 300 mL of pure water and extract the acetamide three times with 500 mL of CH2Cl2. 4. Determine the amount of 20 by absorbance spectrophotometry at l = 260 nm (see Subheading 3.3.2). 5. Characterize the pure tetragalactose oligonucleotide conjugate product 20 by MALDI-TOF mass spectrometry and reversed-phase HPLC (gradient: 6–24% (v/v) CH3CN in 0.05 M TEAAc for 20 min) (Fig. 6). (HPLC RT = 13.83 min, lmax = 262  nm, e = 124,700  L/mol/cm. MALDI-TOF MS (negative mode) m/z: calcd for C211H311N56O116P16 [M − H]−: 5983.615; found 5983.36. Isolated amount: 23 nmol.)

Synthesis of a Glycomimetic Oligonucleotide Conjugate by 1,3-Dipolar Cycloaddition

189

13,83

0

5

10

min

15

20

25

Fig. 6. C18 RP-HPLC profile of the tetragalactosyl-mannose oligonucleotide conjugate 20.

4. Notes 1. Before spotting the reaction mixture, the TLC plates must be neutralized by dipping in a solution of 99:1 (v/v) CH2Cl2/ Et3N and air-dried in order to avoid sample degradation on the plate (due to the acidity of the silica). 2. Before starting the elution, the spot of the reaction mixture must be air-dried to remove pyridine and/or Et3N, which could modify the retention factor (Rf). 3. After the elution, and before the reaction with stain solution, direct visualization using a 254-nm UV lamp can reveal the starting reagents and/or the final compounds. 4. Staining the TLC plate in the stain solution (vanillin or sulfuric acid) reveals the compounds. The plate is slowly heated to remove solvent, dipped in the stain solution, washed with water, stamped with absorbent paper to remove excess of water, and then slowly heated to dry the plate. 5. It is recommended that the chromatography column first be equilibrated with an elution solvent containing Et3N to avoid degradation due to the acidity of the silica. Derivatives bearing a dimethoxytrityl group or phosphoramidite/phosphorodiamidite functions are very sensitive to acidic treatments, so any traces of acid must be neutralized. In our hands, phosphoramidite compounds could not be purified on silica gel columns without first adding Et3N. 6. The Rf values of both compounds are very similar. To overcome this difficulty, it is necessary to elute the TLC plate an

190

Pourceau et al.

additional two or three times after the first elution using the same elution gradient. 7. Before capping the unreacted functions, the loading of the solid support can be determined by a trityl assay as follows: Filter off some of the beads, wash three times with 5 mL of CH2Cl2, dry in a vacuum desiccator over P2O5 for 30  min, and weigh the beads. Calculate the loading of the functionalized solid support as described in step 16. 8. After capping with acetic anhydride in the presence of N-methylimidazole, the loading is typically found to be between 40 and 65 mmol/g of CPG. 9. To prepare these solutions, water must be degassed to avoid degradation of the oligonucleotide due to the oxygen in the presence of copper. Furthermore, it is best to prepare the sodium ascorbate solution just before the reaction to allow the reduction of copper(II) into copper(I) to proceed effectively. The copper sulfate solution can be prepared in advance and stored for several months at −20°C. When both solutions are mixed together, the color turns brown and opaque. 10. We have noticed that a very efficient CuAAC reaction occurs when using microwave (MW) assistance. If no microwave synthesizer is available, the reaction should be performed under vigorous stirring at 70°C for at least 2  h. However, reactions performed without MW assistance are typically less reproducible. 11. While transferring the beads into the column, be careful to ensure that no beads remain on the rim of the column, as this will result in column malfunction and failed reaction steps. 12. It is essential to have the solid support 10 be as dry as possible since trace amounts of water will react with the phosphoramidite 12, leading to less efficient introduction of pentynyl groups on the mannose core. 13. This cycle is derived from the standard elongation cycle that is available in the DNA synthesizer’s program library. The detritylation step must be removed since the mannose hydroxyls are free, but this is also to keep the dimethoxytrityl group present on the solid support for the further synthesis of the oligonucleotide. The coupling time of the phosphoramidite is longer: 180 s instead of 20 s; and a double coupling step is applied to ensure the success of the reaction on the four hydroxyls on the mannose. (Note that three of the hydroxyls are secondary alcohols that are less reactive than the standard 5¢-primary alcohol of a nucleoside.) 14. After the first incorporation, the cycle can be paused and some of the CPG beads can be deprotected (with 1  mL of NH4OH for 15 h at room temperature) to verify the ­successful

Synthesis of a Glycomimetic Oligonucleotide Conjugate by 1,3-Dipolar Cycloaddition

191

incorporation of the modified phosphoramidite 12 by reversed-phase HPLC and MALDI-TOF mass spectrometry before continuing the elongation. This verification process cannot be carried out before the incorporation of a nucleoside because there is no chromophore on the support, and the mass is too low to be detectable by MALDI-TOF MS. 15. Since the present oligonucleotide sequence has no deoxyguanosine protected with an isobutyryl group, the deprotection step can be limited to 2  h. On the other hand, sequences containing dG must be treated for 5 h at 55°C to yield full deprotection. 16. The extinction coefficient (e) of the oligonucleotide at l = 260 nm is calculated as the sum of the extinction coefficients of the constitutive nucleosides: e(dA) = 15,400  L/mol/cm; e(dG) = 11,500  L/mol/cm; e(dC) = 7,400  L/mol/cm; and e(dT) = 8,700 L/mol/cm. 17. To facilitate the purification of the tetragalactose oligonucleotide conjugate, we use the tetraacetylated galactose azide derivative 19 instead of the unprotected galactosyl moiety in order to confer lipophilicity to the resulting oligonucleotide conjugate, thus allowing for an easier RP-HPLC separation step. After RP-HPLC purification, a supplementary deprotection step that involves the hydrolysis of acetyl groups is required. A final extraction with ethyl ether is then performed to remove the acetamide. 18. FTIR is a useful method to characterize azide derivatives since they exhibit a strong vibration at 2,096 cm−1. 19. The dissolution of the crude with cyclohexane can be difficult; add a few milliliters of ethyl acetate to help with the dissolution process. 20. We have noticed that when CuAAC reactions are performed in solution, the use of MW assistance is less important. In this case, however, the reaction is usually incubated for a longer period of time (2–3  h) to obtain a similar yield level. It is essential that the click reaction be completed within 5 h to ensure that there is no significant degradation of the oligonucleotide moiety. References 1. Varki, A. (1993) Biological Roles of Oligosaccharides – All of the Theories Are Correct. Glycobiology 3, 97–130. 2. Dwek, R. A. (1996) Glycobiology: Toward understanding the function of sugars. Chem. Rev. 96, 683–720.

3. Lis, H., and Sharon, N. (1998) Lectins: Carbohydrate-Specific Proteins That Mediate Cellular Recognition. Chem. Rev. 98, 637–674. 4. Imberty, A., Wimmerova, M., Mitchell, E. P., and Gilboa-Garber, N. (2004) Structures of the lectins from Pseudomonas aeruginosa:

192

Pourceau et al.

insights into the molecular basis for host glycan recognition. Microbes Infect. 6, 221–228. 5. Chemani, C., Imberty, A., de Bentzmann, S., Pierre, M., Wimmerova, M., Guery, B. P., and Faure, K. (2009) Role of LecA and LecB Lectins in Pseudomonas aeruginosa-Induced Lung Injury and Effect of Carbohydrate Ligands. Infect. Immun. 77, 2065–2075. 6. Deguise, I., Lagnoux, D., and Roy, R. (2007) Synthesis of glycodendrimers containing both fucoside and galactoside residues and their binding properties to Pa-IL and PA-IIL lectins from Pseudomonas aeruginosa. New J. Chem. 31, 1321–1331. 7. Lundquist, J. J., and Toone, E. J. (2002) The cluster glycoside effect. Chem. Rev. 102, 555–578. 8. Darbre, T., and Reymond, J. L. (2008) Glycopeptide dendrimers for biomedical applications. Curr. Top. Med. Chem. 8, 1286–1293. 9. Singh, Y., Renaudet, O., Defrancq, E., and Dumy, P. (2005) Preparation of a multitopic glycopeptide-oligonucleotide conjugate. Org. Lett. 7, 1359–1362. 10. Marra, A., Moni, L., Pazzi, D., Corallini, A., Bridi, D., and Dondoni, A. (2008) Synthesis of sialoclusters appended to calix[4]arene platforms via multiple azide-alkyne cycloaddition. New inhibitors of hemagglutination and cytopathic effect mediated by BK and influenza A viruses. Org. Biomol. Chem. 6, 1396–1409. 11. Cecioni, S., Lalor, R., Blanchard, B., Praly, J. P., Imberty, A., Matthews, S. E., and Vidal, S. (2009) Achieving high affinity towards a bacterial lectin through multivalent topological isomers of calix[4]arene glycoconjugates. Chem. Eur. J. 15, 13232–13240. 12. Dubber, M., and Lindhorst, T. K. (2001) Trehalose-based octopus glycosides for the synthesis of carbohydrate-centered PAMAM dedrimers and thiourea-bridged glycoclusters. Org. Lett. 3, 4019–4022. 13. Ortega-Munoz, M., Perez-Balderas, F., Morales-Sanfrutos, J., Hernandez-Mateo, F., Isac-Garcia, J., and Santoyo-Gonzalez, F. (2009) Click Multivalent Heterogeneous Neoglycoconjugates – Modular Synthesis and Evaluation of Their Binding Affinities. Eur. J. Org. Chem., 2454–2473. 14. Garcia-Lopez, J. J., Hernandez-Mateo, F., Isac-Garcia, J., Kim, J. M., Roy, R., SantoyoGonzalez, F., and Vargas-Berenguel, A. (1999) Synthesis of per-glycosylated beta-cyclodextrins having enhanced lectin binding affinity. J. Org. Chem. 64, 522–531.

15. Matsuura, K., Hibino, M., Yamada, Y., and Kobayashi, K. (2001) Construction of GlycoClusters by Self-Organization of SiteSpecifically Glycosylated Oligonucleotides and Their Cooperative Amplification of LectinRecognition. J. Am. Chem. Soc. 123, 357–358. 16. Zatsepin, T. S., and Oretskaya, T. S. (2004) Synthesis and applications of oligonucleotidecarbohydrate conjugates. Chem. Biodiversity 1, 1401–1417. 17. Biessen, E. A. L., Vietsch, H., Rump, E. T., Fluiter, K., Kuiper, J., Bijsterbosch, M. K., and Van Berkel, T. J. C. (1999) Targeted delivery of oligodeoxynucleotides to parenchymal liver cells in vivo. Biochem. J. 340, 783–792. 18. Katajisto, J., Heinonen, P., and Lonnberg, H. (2004) Solid-phase synthesis of oligonucleotide glycoconjugates bearing three different glycosyl groups: Orthogonally protected bis (hydroxymethyl)-N,N¢-bis(3-hydroxypropyl) malondiamide phosphoramidite as key building block. J. Org. Chem. 69, 7609–7615. 19. Katajisto, J., Virta, P., and Lonnberg, H. (2004) Solid-phase synthesis of multiantennary oligonucleotide glycoconjugates utilizing on-support oximation. Bioconjugate Chem. 15, 890–896. 20. Bouillon, C., Meyer, A., Vidal, S., Jochum, A., Chevolot, Y., Cloarec, J. P., Praly, J. P., Vasseur, J. J., and Morvan, F. (2006) Microwave assisted “click” chemistry for the synthesis of multiple labeled-carbohydrate oligonucleotides on solid support. J. Org. Chem. 71, 4700–4702. 21. Beaucage, S. L., and Caruthers, M. H. (1981) Deoxynucleoside phosphoramidites – a new class of key intermediates for deoxypolynucleotide synthesis. Tetrahedron Lett. 22, 1859–1862. 22. Tornoe, C. W., Christensen, C., and Meldal, M. (2002) Peptidotriazoles on solid phase: [1,2,3]-triazoles by regiospecific copper(I)catalyzed 1,3-dipolar cycloadditions of terminal alkynes to azides. J. Org. Chem. 67, 3057–3064. 23. Rostovtsev, V. V., Green, L. G., Fokin, V. V., and Sharpless, K. B. (2002) A Stepwise Huisgen Cycloaddition Process: Copper(I)Catalyzed Regioselective “Ligation” of Azides and Terminal Alkynes. Angew. Chem. Int. Ed. 41, 2596–2599. 24. Wacker, R., and Niemeyer, C. M. (2004) DDI-mu FIA – A readily configurable microarray-fluorescence immunoassay based on DNAdirected immobilization of proteins. Chembiochem 5, 453–459.

Synthesis of a Glycomimetic Oligonucleotide Conjugate by 1,3-Dipolar Cycloaddition 25. Chevolot, Y., Bouillon, C., Vidal, S., Morvan, F., Meyer, A., Cloarec, J. P., Jochum, A., Praly, J. P., Vasseur, J. J., and Souteyrand, E. (2007) DNA-based carbohydrate biochips: A platform for surface glyco-engineering. Angew. Chem. Int. Ed. 46, 2398–2402. 26. Pourceau, G., Meyer, A., Vasseur, J. J., and Morvan, F. (2009) Azide Solid Support for 3¢-Conjugation of Oligonucleotides and Their Circularization by Click Chemistry. J. Org. Chem. 74, 6837–6842. 27. Hasegawa, T., Numata, M., Okumura, S., Kimura, T., Sakurai, K., and Shinkai, S. (2007) Carbohydrate-appended curdlans as a new family of glycoclusters with binding properties both for a polynucleotide and lectins. Org. Biomol. Chem. 5, 2404–2412. 28. Pourceau, G., Meyer, A., Vasseur, J. J., and Morvan, F. (2009) Synthesis of Mannose and

193

Galactose Oligonucleotide Conjugates by Bi-click chemistry. J. Org. Chem. 74, 1218–1222. 29. Pourceau, G., Meyer, A., Vasseur, J. J., and Morvan, F. (2008) Combinatorial and automated synthesis of phosphodiester galactosyl cluster on solid support by click chemistry assisted by microwaves. J. Org. Chem. 73, 6014–6017. 30. Lonnberg, H. (2009) Solid-Phase Synthesis of Oligonucleotide Conjugates Useful for Delivery and Targeting of Potential Nucleic Acid Therapeutics. Bioconjugate Chem. 20, 1065–1094. 31. Pon, R. T., and Yu, S. Y. (1997) HydroquinoneO,O¢-diacetic acid (Q-linker) as a replacement for succinyl and oxalyl linker arms in solid phase oligonucleotide synthesis. Nucleic Acids Res. 25, 3629–3635.

Chapter 12 Site-Specific DNA Labeling by Staudinger Ligation Samuel H. Weisbrod, Anna Baccaro, and Andreas Marx Abstract Site-specific and chemoselective labeling of DNA is still a difficult task. The Staudinger ligation is a bioorthogonal reaction between azides and phosphines that requires no catalyst to proceed, allowing for mild reaction conditions. The reaction may be extended for site-specific labeling of DNA using azidomodified triphosphates, which can be incorporated site-specifically into DNA strands by DNA polymerases in a template-dependent manner. The azido-modified DNA, in turn, can be labeled by suitable phosphines. This protocol describes (1) the synthesis of an azido-TTP analogue; (2) the enzymatic synthesis of azido-modified DNA; (3) the synthesis of suitable phosphine labels; and (4) the labeling of azido-DNA with biotin–phosphine by Staudinger ligation with approximately 70% conversion. Key words: Site-specific DNA labeling, DNA modification, Staudinger ligation, Phosphine, Modified triphosphate, Azido thymidine, DNA polymerase, Azido-DNA, Biotin

1. Introduction Several conjugation reactions have been used to label DNA sitespecifically; however, truly chemoselective methods are rarely represented (1). Besides Diels Alder reactions and copper-catalyzed azide alkyne 1,3 dipolar cycloadditions, the Staudinger ligation is a type of chemoselective conjugation reaction that has been used for site-specific labeling of DNA (2). The Staudinger ligation reaction occurs between an azide and a phosphine to form an azaylide that can be trapped by an acyl group to form a stable amide bond (3). Since both functionalities exhibit high bioorthogonality and no further reagents are required, the Staudinger ligation has found several in  vivo applications (4) and should allow the attachment of a wide variety of functional labels. Since the direct incorporation of an azido or phosphor(III) functionality by standard phosphoramidite chemistry is not

Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_12, © Springer Science+Business Media, LLC 2011

195

196

Weisbrod, Baccaro, and Marx

Fig. 1. Scheme for labeling DNA with biotin by Staudinger ligation. 5-(5-azido-1-pentynyl)-2¢-deoxyuridine (azido-TTP) is first incorporated by a DNA polymerase instead of TTP in a growing DNA strand. The azido label is then reacted with a suitable phosphine label. In this example, biotin–phosphine is attached to the DNA strand.

a­ pplicable, our approach involves the incorporation of azidofunctionalized deoxynucleoside triphosphates by a DNA polymerase and subsequent labeling of enzymatically generated DNA by Staudinger ligation using phosphine labels without by-product formation (see ref. 5 and Fig. 1). In this chapter, the site-specific labeling of short enzymatically generated DNA strands with biotin will be used to illustrate the method.

2. Materials 2.1. Synthesis of Azido-TTP

1. TLC plates Silica gel 60 F254 (Merck, Darmstadt, Germany). 2. Imidazole. 3. tert-Butyldimethylsilyl chloride. 4. 2¢-Deoxy-5-iodouridine. 5. Triethylamine. 6. Pent-4-yn-1-ol. 7. Tetrakis(triphenylphosphine)palladium(0). 8. Copper(I)iodide. 9. N,N-Diisopropylethylamine.

Site-Specific DNA Labeling by Staudinger Ligation

197

10. Methanesulfonyl chloride. 11. Sodium azide. 12. Tetrabutylammonium fluoride solution: 1  M tetrabutylammonium fluoride in THF. 13. Acetic acid. 14. TEAB buffer: 1  M triethylamine is saturated with gaseous CO2 (from dry ice) overnight, pH 7.5. 15. TEAA buffer: 50  mM triethylamine, 50  mM acetic acid, pH 7.0. 16. 1,8-Bis(dimethylaminonaphthalene). 17. Phosphoryl chloride. 18. Trimethyl phosphate. 19. Tri-n-butylamine. 20. Bis (see preparation)-tri-n-butylammonium pyrophosphate (prepared as described in Subheading 3.1.6.1). 21. Amberlite® IR120, H+-resin. 22. Diethylaminoethyl (DEAE) Sephadex® A-25. 23. Reversed-phase medium-pressure liquid chromatography (MPLC) column: Lobar® LiChroprep® RP-18 (Merck). 2.2. Synthesis of Phosphine Labels

1. 1-Methyl 2-iodoterephtalate. 2. Palladium(II) acetate. 3. Diphenylphosphine. 4. Carbonyldiimidazole. 5. N-Boc-4,7,10-trioxa-1,13-tridecanediamine. 6. Hydrogen chloride solution: 4  M hydrogen chloride in dioxane. 7. Biotin.

2.3. Enzymatic Incorporation of Azido-TTP ( 6) into DNA

1. Deoxyribonucleotide-5¢-triphosphates: dATP (100  mM); dCTP (100 mM); dGTP (100 mM); and dTTP (100 mM) (MBI Fermentas, St. Leon-Roth, Germany). 2. Oligonucleotides: Primer: 5¢-GAC CCA CTC CAT CGA GAT TTC TC-3¢ (10  mM); Template: 5¢-GCG CTG GCA CGG GAG AAA TCT CGA TGG AGT GGG TC-3¢ (10 mM) (Metabion, Martinsried, Germany). 3. Pwo DNA polymerase (1  U/ml) (Peqlab, Erlangen, Germany). 4. 210 mM ethylenediaminetetraacetic acid (EDTA) solution. 5. illustra MicroSpinTM G-25 columns (GE Healthcare, Freiburg, Germany).

198

Weisbrod, Baccaro, and Marx

2.4. Labeling of Azido-DNA by Staudinger Ligation

1. Sodium carbonate–bicarbonate buffer solution: 1 M Na2CO3– NaHCO3, pH 9.0.

3. Methods The methods described below outline (1) the chemical synthesis of an azido-functionalized thymidine triphosphate analogue (Fig. 2); (2) the synthesis of appropriate phosphines for Staudinger ligation reaction; (3) enzymatic generation of azido-modified DNA; and (4) the site-specific labeling of azido-modified DNA by Staudinger ligation. 3.1. Synthesis of Azido-TTP

Water-sensitive chemical reactions are carried out in oven-dried glassware under an argon atmosphere. Anhydrous solvents are stored over molecular sieves under argon. Thin-layer ­chromatography

Fig. 2. Synthesis of 5-(5-azido-1-pentynyl)-2¢-deoxyuridine (azido-TTP).

Site-Specific DNA Labeling by Staudinger Ligation

199

(TLC) is performed with alumina plates, and compounds are visualized by UV light (254 nm). Column chromatographic separations of compounds are achieved by using 50–100 times more silica gel than sample material. NMR spectra are recorded with CDCl3 (d = 7.24) or CD3OD (dCD3 = 3.31) as solvent. Electrospray ionization mass spectra (ESI–MS) are obtained in positive or negative ion mode with the samples dissolved in methanol or methanol/water (1:1, v/v). The synthesis steps below are based on a published procedure (6). 3.1.1. 3 ¢,5 ¢-Bis-O(tert-Butyldimethylsilyl)-2 ¢Deoxy-5-Iodouridine (1)

1. Add imidazole (0.96 g, 14.1 mmol) to a stirred solution of tert-butyldimethylsilyl chloride (1.28 g, 8.5 mmol) in anhydrous DMF (6 mL) at 0°C. 2. Add 2¢-deoxy-5-iodouridine (1 g, 2.82 mmol) at 0°C and stir 3 h at room temperature (RT). 3. Add ice-cold water (100 mL) and extract with diethyl ether (3 × 100 mL). Combine the organic phases, dry with magnesium sulfate and remove the solvent under reduced pressure. 4. Purify the crude product by column chromatography (silica gel, EtOAc-hexane, 2:3). Nucleoside 1 (1.61  g, 97%) is obtained as a white foam (Rf = 0.3, EtOAc-hexane, 1:2); 1H NMR (400  MHz, CDCl3): d 0.07, 0.08, 0.15, 0.16 (4s, 12H, 2 (CH3)2Si), 0.88, 0.91 (2s, 18H, 2 (CH3)3CSi), 1.99 (ddd, J = 5.9, 8.1, 13.3 Hz, 1H, 2¢-Ha), 2.30 (ddd, J = 2.2, 5.6, 13.3 Hz, 1H, 2¢-Hb), 3.76 (dd, J = 2.3, 11.5 Hz, 1H, 5¢-Ha), 3.89 (dd, J = 2.3, 11.5  Hz, 1H, 5¢-Hb), 3.99 (d, J = 2.3 Hz, 1H, 4¢-H), 4.38–4.40 (m, 1H, 3¢-H), 6.27 (dd, J = 5.6, 8.1 Hz, 1H, 1¢-H), 8.09 (s, 1H, 6-H), 8.55 (br s, 1H, NH).

3.1.2. 3 ¢,5 ¢-Bis-O(tert-Butyldimethylsilyl)-5(5-Hydroxy-1-Pentynyl)-2¢Deoxyuridine (2)

1. Add triethylamine (482  mL, 3.43  mmol), pent-4-yn-1-ol (476 mL, 5.16 mmol) and tetrakis(triphenylphosphine)palladium(0) (27  mg, 0.17  mmol) to a stirred suspension of nucleoside 1 (1 g, 1.72 mmol) and copper(I)iodide (65 mg, 0.34 mmol) in anhydrous DMF (5 mL) (see Note 1). 2. Stir the orange solution at RT for 16 h until no more starting material is detected by TLC. 3. Add 10 mL of saturated sodium bicarbonate solution. 4. Extract with ethyl acetate (3 × 10 mL), combine the organic phases, dry the organic phase with magnesium sulfate and remove the solvent under reduced pressure. 5. Purify the crude product by column chromatography (silica gel, EtOAc-hexane, 1:1 to 3:2, v/v). Nucleoside 2 (750 mg, 81%) is obtained as an off-white solid (Rf = 0.3, EtOAc-hexane, 3:2); 1H NMR (600  MHz, CDCl3): d = 0.09, 0.1, 0.15, 0.16 (4s, 12H, 2 (CH3)2Si), 0.91, 0.95

200

Weisbrod, Baccaro, and Marx

(2s, 18H, 2 (CH3)3CSi), 1.77–1.86 (m, 2H, CH2CH2CH2OH), 2.00–2.07 (m, 1H, H-2¢a), 2.31 (ddd, J = 13.1, 5.7, 2.5 Hz, 1H, H-2¢b), 2.51 (t, J = 6.8  Hz, 2H, CH2CH2CH2OH), 3.76–3.81 (m, 3H, H-5¢b, CH2CH2CH2OH), 3.91 (dd, J = 11.4, 2.0  Hz, 1H, H-5¢a), 3.97–3.99 (m, 1H, H-4¢), 4.40–4.43 (m, 1H, H-3¢), 6.30 (t, J = 5.7, 5.7 Hz, 1H, H-1¢), 7.93 (s, 1H, H-6), 8.28 (br s, 1H, NH); ESI–MS: m/z = 561.5 [M + Na]+. 3.1.3. 3 ¢,5 ¢-Bis-O-(tertButyldimethylsilyl)-5(5-Methanesulfonatepent1-ynyl)-2 ¢-Deoxyuridine (3)

1. Add N,N-diisopropylethylamine (182 mL, 1.07 mmol) to a stirred solution of nucleoside 2 in anhydrous CH2Cl2 (5 mL) at 0°C and stir for 30 min at 0°C. 2. Add methanesulfonyl chloride (55 mL, 0.71 mmol) dropwise and stir for 40 min at 0°C. 3. Add saturated sodium bicarbonate solution (10  mL) and extract with diethyl ether (3 × 15  mL). Wash the combined phase with brine (15 mL), dry with magnesium sulfate, and remove the solvent under reduced pressure. 4. Purify the crude product by column chromatography (silica gel, EtOAc-hexane, 1:2). Nucleoside 3 (270  mg, 74%, see Note 2) is obtained as a white foam (Rf = 0.8, EtOAc-hexane, 3:1); 1H NMR (600 MHz, CDCl3): d = 0.07, 0.08, 0.13, 0.14 (4s, 12H, 2 (CH3)2Si), 0.89, 0.93 (2s, 18H, 2 (CH3)3CSi), 1.97–2.05 (m, 3H, H-2¢a, CH2CH2CH2OMs), 2.31 (ddd, J = 13.1, 6.0, 2.7 Hz, 1H, H-2¢b), 2.55 (dd, J = 6.8, 6.8 Hz, 2H, CH2CH2CH2OMs), 3.07 (s, 3H, OSO2CH3), 3.77 (dd, J = 11.5, 2.1  Hz, 1H, H-5¢b), 3.89 (dd, J = 11.5, 2.4 Hz, 1H, H-5¢a), 3.96–3.98 (m, 1H, H-4¢), 4.37– 4.41 (m, 3H, H-3¢, CH2CH2CH2OMs), 6.27 (dd, J = 7.3, 6.0 Hz, 1H, H-1¢), 7.94 (s, 1H, H-6), 8.05 (br s, 1H, NH); ESI–MS: m/z = 639.4 [M + Na]+.

3.1.4. 3 ¢,5 ¢-Bis-O(tert-Butyldimethylsilyl)5-(5-Azido-1-Pentynyl)-2 ¢Deoxyuridine (4)

1. Add sodium azide (46 mg, 0.71 mmol) to a stirred solution of nucleoside 3 (70  mg, 0.11  mmol) in anhydrous DMF (2 mL) and stir for 20 h at 35°C. 2. Add saturated sodium bicarbonate solution (5  mL) and extract with ethyl acetate (3 × 5  mL). Dry the combined organic phases with magnesium sulfate, and remove the solvent under reduced pressure. 3. Purify the crude product by column chromatography (silica gel, EtOAc-hexane, 1:3 to 1:1). Nucleoside 4 (61  mg, 95%) is obtained as a white foam (Rf = 0.8, EtOAc-hexane, 1:1); 1H NMR (600  MHz, CDCl3): d = 0.09, 0.1, 0.15, 0.16 (4s, 12H, 2 (CH3)2Si), 0.91 0.95 (2s, 18H, 2 (CH3)3CSi), 1.85 (dddd, J = 6.8, 6.8, 6.8, 6.8 Hz, 2H, CH2CH2CH2N3), 2.03 (ddd, J = 13.3, 7.7, 6.3 Hz, 1H, H-2¢a), 2.31 (ddd, J = 13.3, 5.9, 2.7 Hz, 1H, H-2¢b), 2.51 (t, J = 6.8 Hz,

Site-Specific DNA Labeling by Staudinger Ligation

201

2H, CH2CH2CH2N3), 3.45 (t, J = 6.8 Hz, 2H, CH2CH2CH2N3), 3.78 (dd, J = 11.5, 2.2  Hz, 1H, H-5¢b), 3.91 (dd, J = 11.5, 2.3 Hz, 1H, H-5¢a), 3.98 (ddd, J = 2.2, 2.3, 2.2 Hz, 1H, H-4¢), 4.40–4.43 (m, 1H, H-3¢), 6.3 (dd, J = 7.7, 5.9 Hz, 1H, H-1¢), 7.93 (s, 1H, H-6), 8.27 (br s, 1H, NH); ESI–MS: m/z = 586.6 [M + Na]+. 3.1.5. 5-(5-Azido1-Pentynyl)-2 ¢Deoxyuridine (5)

1. Add tetrabutylammonium fluoride (1  M in THF, 226  mL, 0.23  mmol) to a stirred solution of nucleoside 4 (58  mg, 0.10 mmol) in anhydrous THF (2 mL) at 0°C. 2. Stir and let warm to RT for 16 h. 3. Remove the solvent under reduced pressure. 4. Purify the crude product by column chromatography (silica gel, CH2Cl2–MeOH, 20:1). Nucleoside 5 (32  mg, 93%) is obtained as a white foam (Rf = 0.4, CH2Cl2–MeOH, 20:1); 1H NMR (400 MHz, CD3OD): d = 1.75 (dddd, J = 6.8, 6.8, 6.8, 6.8 Hz, 2H, CH2CH2CH2N3), 2.10–2.26 (m, 2H, H-2¢a, H-2¢b), 2.43 (t, J = 6.8  Hz, 2H, CH2CH2CH2N3), 3.40 (t, J = 6.8  Hz, 2H, CH2CH2CH2N3), 3.67 (dd, J = 12.0, 3.4  Hz, 1H, H-5¢b), 3.75 (dd, J = 12.0, 3.0 Hz, 1H, H-5¢a), 3.84–3.88 (m, 1H, H-4¢), 4.31–4.36 (m, 1H, H-3¢), 6.18 (t, J = 6.6, 6.6 Hz, 1H, H-1¢), 8.19 (s, 1H, H-6); ESI–MS: m/z = 357.9 [M + Na]+.

3.1.6. 5-(5-Azido1-Pentynyl)-2 ¢Deoxyuridine-5 ¢Triphosphate (6)

1. Prepare a solution of tri-n-butylamine (4.77 mL, 20 mmol) in ethanol (40 mL) and cool with ice (solution A).

3.1.6.1. Preparation of Bis (see preparation)Tri-n-Butylammonium Pyrophosphate

3. Dissolve tetrasodium diphosphate decahydrate (4.46  g, 10 mmol) in water (75 mL) and load it onto the Amberlite® column.

2. Fill a column (20 cm × 3 cm) with fresh Amberlite® IR-120 (H+ form) and wash with water until pH 5.

4. Elute the column dropwise with water directly into the icecold stirred solution A until the eluate again reaches pH 5. 5. Remove the solvent under reduced pressure at 30°C. Co-evaporate the residue with anhydrous ethanol and anhydrous DMF (each 3 × 10 mL). Attention: do not heat above 30°C for the evaporation step. 6. Dissolve the residue in anhydrous DMF (total volume 20 mL) and store over molecular sieves (4 Å) at 4°C.

3.1.7. Preparation of Triphosphate (6)

1. Dry in vacuo 1,8-bis(dimethylaminonaphthalene) (28  mg, 0.13 mmol) and nucleoside 5 (29 mg, 86 mmol) overnight in the dark (see Note 3). 2. Dissolve both reagents in anhydrous trimethyl phosphate (1 mL) and cool to 0°C.

202

Weisbrod, Baccaro, and Marx

3. Add dropwise freshly distilled phosphoryl chloride (10.5 mL, 0.11 mmol) and stir for 3 h at 0°C (see Note 4). 4. Simultaneously add 0.5  M Bis (see preparation)-tri-n-butylammonium pyrophosphate in anhydrous DMF (0.86 mL, 0.43 mmol) and tri-n-butylamine (229 mL, 0.86 mmol), and stir for 10 min. 5. Quench with 1 M TEAB buffer (5 mL) and wash the aqueous phase with ethyl acetate (2 × 10  mL). Lyophilize the aqueous phase. 6. Purify the residue by ion-exchange chromatography (DEAESephadex® A-25 (18 g dry)) with a linear gradient of TEAB buffer (0.1–1 M, total volume of 1,500 mL, 2 mL/min flow rate at 4°C). Lyophilize the product-containing fractions (check by ESI–MS, negative mode). 7. Purify by reversed-phase medium-pressure liquid chromatography (RP-MPLC) (RP-18, 40–63  mm) using a gradient of 5% (200 ml), 20% (200 ml) and 40% (200 ml) acetonitrile in 50 mM (TEAA buffer). The triphosphate elutes at 20% acetonitrile in 50 mM TEAA buffer. Lyophilize several times. 8. Store the triphosphate as a 100  mM solution in water at −20°C (see Note 5). Nucleotide 6 (26 mg, 30%) is obtained as an oil; 1H NMR (400  MHz, CD3OD): d = 1.20–1.35 [m, 36H, N(CH2CH3)3], 1.8–1.93 (m, 2H, CH2CH2CH2N3), 2.30–2.40 (m, 2H, H-2¢b, H-2¢a), 2.47–2.55 (m, 2H, CH2CH2CH2N3), 3.13–3.25 [m, H-5¢a, H-5¢b, N(CH2CH3)3], 3.25–3.46 (m, 2H, CH2CH2CH2N3), 4.1–4.2 (m, 1H, H-3¢), 4.56–4.62 (m, 1H, H-4¢), 6.26 (m, 1H, H-1¢), 8.0 (s, 1H, H-6); 31P NMR (162 MHz, MeOD): d = −22.5 (m, 1 P, Pb), −10.6 (d, J = 20.2 Hz, 1 P, Pa), −9.3 (d, J = 20.2 Hz, 1 P, Pg); ESI–MS: m/z = 575.0 [M–H]−. 3.2. Synthesis of Biotin–Phosphine Label (Fig. 3)

Additionally to the remarks in Subheading 3.1, it should be mentioned that the described phosphines are to some extent susceptible to oxidation. As a precaution, store the phosphines under nitrogen or argon at −20°C and use (whenever possible) degassed solutions for the synthesis procedures. Phosphine oxides exhibit 31 P NMR signals at d = 30–40 ppm.

3.2.1. 1-Methyl 2-Diphenylphosphi­ noterephtalate ( 7)

The synthesis is based on a literature procedure (3). 1. Dissolve 1-methyl 2-iodoterephtalate (1.73  g, 5.66  mmol) and palladium(II)acetate (30 mg, 0.13 mmol) in anhydrous methanol (18 mL). 2. Add triethylamine (1.57 mL, 11.32 mmol) and diphenylphosphine (0.98  mL, 5.66  mmol) and reflux the solution overnight.

Site-Specific DNA Labeling by Staudinger Ligation

203

Fig. 3. Synthesis of the biotin–phosphine label.

3. Remove the solvent under reduced pressure, mix the residue with CH2Cl2 and water (200 mL each), separate phases, wash the organic phase with hydrochloric acid (1 M, 100 mL), dry the organic phase with magnesium sulfate, and remove the solvent under reduced pressure. 4. Purify the crude product by column chromatography (silica gel, CH2Cl2–MeOH, 9:1). Phosphine 7 (0.99  g, 48%) is obtained as a yellow solid (Rf = 0.3, CH2Cl2–MeOH, 9:1); 1H NMR (400 MHz, CDCl3): d = 3.68 (s, 3H, CH3), 7.19–7.30 (m, 10H, aryl), 7.58–7.61 (m, 1H, H-3), 7.99–8.06 (m, 2H, H-5, H-6); 31P NMR (CDCl3) d = −3.2. 3.2.2. 1-Methyl 2-Diphenylphosphino4-(N-Boc-4,7,10-Trioxa13-Tridecaneamine) Carbamoyl Benzoate ( 8)

1. Dissolve phosphine 7 (200  mg, 0.55  mmol) in anhydrous CH2Cl2 (6 mL) and anhydrous DMF (2 mL) and stir at RT. 2. Add carbonyldiimidazole (98 mg, 0.6 mmol) and stir for 1 h. 3. Add a solution of N-Boc-4,7,10-trioxa-1,13-tridecanediamine (176 mg, 0.55 mmol) in anhydrous CH2Cl2 (2 mL) and stir for 2 h. 4. Wash the reaction mixture with saturated ammonium chloride solution (10 mL), dry the organic phase with magnesium sulfate and remove the solvent under reduced pressure. 5. Purify the crude product by column chromatography (silica gel, CH2Cl2 to CH2Cl2–MeOH, 19:1).

204

Weisbrod, Baccaro, and Marx

Phosphine 8 (320  mg, 87%) is obtained as a yellow oil (Rf = 0.4, CH2Cl2–MeOH, 19:1); 1H NMR (400 MHz, CDCl3): d = 1.39 (s, 9H, (CH3)3C), 1.67 (p, J = 6 Hz, 2H, CH2), 1.75 (p, J = 6 Hz, 2H, CH2), 3.14 (m, 2H, NCH2), 3.45–3.40 (m, 6H, OCH2, NCH2) 3.45–3.58 (m, 8H, OCH2), 3.69 (s, 3H, CH3), 7.23–7.30 (m, 10H, aryl), 7.37 (dd, J = 3.6, 1.2 Hz, 1H, H-3), 7.71 (dd, J = 8, 1.2 Hz, 1H, H-6), 8.01 (dd, J = 8.0, 3.6 Hz, 1H, H-5); 31P NMR (CDCl3) d = −3.2; ESI–MS: m/z = 667.1 [M + H]+. 3.2.3. 1-Methyl 2-Diphenylphosphino-4(4, 7, 10-Trioxa-13Aminotridecyl) Carbamoylbenzoate (9)

1. Stir phosphine (8) (320  mg, 0.48  mmol) for 2  h at RT in hydrogen chloride in dioxane (5 mL). 2. Remove the solvent under reduced pressure. 3. Purify the crude product by column chromatography (silica gel, CH2Cl2–MeOH, 9:1) (see Note 6). Phosphine 9 (198  mg, 73%) is obtained as a yellow solid (Rf = 0.35, CH2Cl2–MeOH, 9:1); 1H NMR (400 MHz, CDCl3): 1.69–1.81 (m, 2H, CH2), 1.84–1.96 (m, 2H, CH2), 3.04 (m, 2H, NCH2), 3.30–3.6 (m, 8H, NH2, OCH2, NCH2), 3.50 (m, 8H, OCH2) 3.66 (s, 3H, CH3), 7.19–7.30 (m, 10H, aryl), 7.43 (dd, J = 3.7, 1.6  Hz, 1H, H-3), 7.74 (t, J = 5.6  Hz, 1H, NH), 7.91 (dd, J = 8.1, 1.6 Hz, 1H, H-6), 8.00 (dd, J = 8.1, 3.7 Hz, 1H, H-5); 31P NMR (CDCl3) d = −3.0; ESI–MS: m/z = 567.0 [M + H]+.

3.2.4. 1-Methyl 2-Diphenylphosphino-4(4, 7, 10-Trioxa-13Biotinamidotridecyl) Carbamoyl Benzoate (10)

1. Stir carbonyldiimidazole (26 mg, 0.16 mmol) and biotin (32 mg 0.13 mmol) in anhydrous DMF (3 mL) for 2 h at RT. 2. Add phosphine 9 (50 mg, 0.09 mmol) in DMF (1 mL) to the reaction mixture and stir overnight at RT. 3. Remove the solvent under reduced pressure. 4. Add water (5 mL), extract with CH2Cl2 (2 × 5 mL), combine the organic phases, dry the organic phase with magnesium sulfate and remove the solvent under reduced pressure. 5. Purify the crude biotin–phosphine product 10 by column chromatography (silica gel, CH2Cl2–MeOH, 94:6). Biotin–phosphine 10 (33  mg, 47%) is obtained as a yellow solid (Rf = 0.25, CH2Cl2–MeOH, 94:6); 1H NMR (400  MHz, CDCl3): 1.40 (p, J = 6.4  Hz, 2H, CH2), 1.57–1.66 (m, 4H, CH2), 1.72 (p, J = 6.4 Hz, 2H, CH2), 1.79 (p, J = 6.4 Hz, 2H, CH2), 2.15 (t, J = 7.6 Hz, 2H, CH2), 2.70 (d, J = 12.8 Hz, 1H, CH), 2.86 (dd, J = 12.8, 4.8 Hz, 1H, CH), 3.08–3.12 (m, 1H, CH), 3.28 (q, J = 6.4  Hz, 2H, NCH2), 3.42–3.60 (m, 14H, OCH2, NCH2), 3.72 (s, 3H, CH3), 4.25 (dd, J = 7.6, 5.2 Hz, 1H, CH), 4.45 (dd, J = 7.6, 4.8  Hz, 1H, CH), 5.61 (s, 1H, NH), 6.41 (s, 1H, NH), 6.74 (t, J = 5.6 Hz, 1H, NH), 7.23 (t, J = 5.6 Hz, 1H, NH), 7.25–7.34 (m, 10H, aryl), 7.42 (dd, J = 3.6, 1.0 Hz,

Site-Specific DNA Labeling by Staudinger Ligation

205

1H, H-3), 7.78 (dd, J = 8.0, 1.0 Hz, 1H, H-6), 8.05 (dd, J = 8.0, 3.6  Hz, 1H, H-5); 31P NMR (CDCl3) d = −3.0; ESI–MS: m/z = 814.9 [M + Na]+. 3.3. Enzymatic Incorporation of Azido-TTP ( 6) into DNA

Incorporation of the azido triphosphate is illustrated with the primer/template duplex listed in Subheading 2.3. For each occurrence of a nonhybridized “A” in the primer/template duplex, azido-TTP (6) is incorporated for later labeling of the synthesized duplex DNA molecule. In this example protocol, a single azidoTTP is incorporated. Extinction coefficients for the single DNA strands at 260  nm are calculated according to Sambrook and Russell (8) and used to adjust the concentration of the oligonucleotide solutions. For radioactive detection on 12% polyacrylamide gels, the primer strand contained approximately 5% 5¢ 32 P-labeled primer (see Note 7). 1. Prepare a solution (dNTP mixture) containing 2.5 mM dATP, 2.5 mM dCTP, 2.5 mM dGTP, and 2.5 mM azido-TTP (6) in water. 2. Prepare the primer/template duplex by adding primer (2.4 mL), template (3.2 mL) and Pwo reaction buffer (2 mL) to water (9.4 mL) in an Eppendorf tube. 3. Anneal the primer/template DNA strands by heating the Eppendorf tube to 95°C for 5 min and let cool down to RT. Cool the solution further to 0°C (see Note 8). 4. Add the dNTP mixture (2  mL) and Pwo DNA polymerase (1 mL) at 0°C to bring the total reaction volume to 20 mL. Heat to 72°C for 30 min and cool down again to 0°C. 5. Quench the reaction with EDTA solution (1  mL). Remove buffers and excess nucleotides using Microspin™ G-25 columns. The formation of azido-modified DNA can be analyzed by denaturing PAGE with 32P-labeled DNA (Fig. 4).

3.4. Labeling of Azido-DNA by Staudinger Ligation

1. Combine the azido-DNA solution (5  mL, as prepared in Subheading 3.3), Na2CO3/NaHCO3 buffer (2.5 mL) and the desired phosphine label (e.g., biotin–phosphine 10) (2.5 mL, 10 mM in DMF). 2. Heat for 12 h at 60°C (see Note 9). 3. The formation of Staudinger ligation products can be visualized by denaturing PAGE with 32P-labeled DNA (Fig. 4).

4. Notes 1. In order to obtain high yields, exclusion of air and moisture is very important for this reaction. Copper(I) iodide dissolves upon addition of triethylamine. Sometimes catalyst residues

206

Weisbrod, Baccaro, and Marx

Fig. 4. Analysis of synthesis reaction products by denaturing polyacrylamide gel electrophoresis. (a) Enzymatic synthesis of azido-modified DNA: The incorporation of azido-TTP (lane 4) in comparison to TTP (lane 3), without TTP (lane 2), and only the primer/template complex (lane 1) as marker. (b) Staudinger ligation with azido-modified DNA and a biotin–phosphine label.

cannot be removed by normal column chromatography. However, analytically pure compounds can be obtained by RP-columns with linear gradients from water to acetonitrile. 2. Nucleoside 3 should not be stored over long time periods. It is best to use it as quickly as possible in the next step. Alternatively, column chromatographic purification can be omitted and the crude nucleoside can be used directly in the next step. In this case, however, we have observed a lower overall yield. 3. For the triphosphate synthesis procedure, it is very important that all reagents and solvents are clean and dry. Trimethyl phosphate is distilled in vacuo and stored over molecular sieves (4 Å), and phosphoryl chloride is either freshly distilled or distilled and stored under argon. 4. The reaction may be monitored with TLC (n-propanol/ water/conc. ammonia, 3:1:1 v/v/v). Occasionally, it is necessary to add more phosphoryl chloride, but do not add more than 0.5 additional equivalents since twofold phosphorylation should be avoided. 5. The concentration is determined by UV absorption using e = 9,000 M−1 cm−1 (287 nm) of compound 2 (7).

Site-Specific DNA Labeling by Staudinger Ligation

207

6. This compound can be stored over long time periods under argon at −20°C and utilized for the attachment of different labels. 7. Appropriate permissions to work with 32P should be obtained. Radioactively labeled substances should be handled with appropriate safety precautions. As an alternative to using radioactive labels, one may also consider using different detection techniques such as silver staining. Primer extension and Staudinger ligation do not rely on the use of one specific type of detection technique. 8. Using a PCR thermocycler, the following cooling procedure can be carried out in a few minutes: Heat to 95°C for 4 min, then cool down stepwise for 30 s at 65, 55, 40, 25, 20°C and hold at 0°C. 9. Staudinger ligation with biotin–phosphine 10 seems to be very robust towards different reaction conditions. Lower temperatures with elongated reaction times are possible as well as reactions in water (~pH 7) or sodium hydroxide solution (pH 12), resulting in similar yields. The effect of cosolvents has also been investigated: In addition to DMF, ethanol appears to be a suitable co-solvent; however, most important is to achieve good solubility of the phosphine in the organic co-solvent mixture. References 1. Niemeyer, C. M. ed. (2004) Bioconjugation Protocols (Methods in Molecular Biology Series Vol. 283). Humana Press, Totowa, NJ. 2. Weisbrod, S. H. and Marx, A. (2008) Novel strategies for the site-specific covalent labeling of nucleic acids. Chem. Commun. 44, 5675–5685. 3. Saxon, E. and Bertozzi, C. R. (2000) Cell surface engineering by a modified Staudinger reaction. Science 287, 2007–2010. 4. Prescher, J. A., Dube D. H. and Bertozzi C. R. (2004) Chemical remodelling of cell surfaces in living animals. Nature 430, 873–877. 5. Weisbrod, S. H. and Marx, A. (2007) A nucleoside triphosphate for site-specific labeling of

DNA by the Staudinger ligation. Chem. Commun. 18, 1828–1830. 6. Baccaro, A. Weisbrod, S. H. and Marx, A. (2007) DNA conjugation by the Staudinger ligation: new thymidine analogues. Synthesis 13, 1949–1954. 7. Robins, M. J. and Barr, P. J. (1983) Nucleic acid related compounds. 39. Efficient conversion of 5-iodo to 5-alkynyl and derived 5-substituted uracil bases and nucleosides. J. Org. Chem. 48, 1854–1862. 8. Sambrook J., Russell D. W. (2001) Molecular Cloning: A Laboratory Manual. Laboratory Press, Cold Spring Harbor, NY.

Chapter 13 Improved Cellular Uptake of Antisense Peptide Nucleic Acids by Conjugation to a Cell-Penetrating Peptide and a Lipid Domain Takehiko Shiraishi and Peter E. Nielsen Abstract Unaided cellular uptake of RNA interference agents such as antisense oligonucleotides and siRNA is extremely poor, and in vivo bioavailability is also limited. Thus, effective delivery strategies for such potential drugs are in high demand. Recently, a novel approach using a class of short cationic peptides known as cell-penetrating peptides (CPPs) is attracting wide attention for a variety of biologically active molecules. CPP-mediated delivery is typically based on the covalent conjugation of the (therapeutic) cargo to CPPs, and is particularly relevant for the delivery of noncharged RNA interference agents such as peptide nucleic acids (PNAs) and morpholino oligomers. Although chemical conjugation to a variety of CPPs significantly improves the cellular uptake of PNAs, the bioavailability (and hence antisense activity) of CPP–PNA ­conjugates is still highly limited by endocytotic entrapment. We have found, however, that this low ­bioavailability can be significantly improved by chemical conjugation to a lipid domain (“Lip,” such as a fatty acid), thereby creating “CatLip”-conjugates. The cellular uptake of these conjugates is conveniently evaluated using a sensitive cellular assay system based on a splicing correction of a mutated luciferase gene in HeLa pLuc705 cells by targeting antisense oligonucleotides to a cryptic splice site. Further improvement in the delivery of CatLip–PNA conjugates is achieved by using auxiliary agents/treatments (e.g., chloroquine, calcium ions, or photosensitizers) to induce endosomal disruption. Key words: Antisense oligonucleotides, Cellular uptake, Cell-penetrating peptide, Catlip ­conjugates, Peptide nucleic acid

1. Introduction Synthetic oligonucleotides, such as aptamers, siRNAs, antisense oligonucleotides, and ribozymes are widely exploited in drug discovery studies. In the case of RNA interference agents (antisense oligonucleotides, siRNA, etc.) their target molecules are located intracellularly (in the nucleus and/or cytosol) and therefore these Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_13, © Springer Science+Business Media, LLC 2011

209

210

Shiraishi and Nielsen

agents need to be delivered into the cells by penetrating the plasma membrane. However, such oligonucleotide-based agents typically have a negligible ability to pass through the cell ­membrane because of their size, hydrophilic properties and not least polyanionic character (1). In order to improve the cellular uptake of oligonucleotides, several chemical derivatization, conjugation, and complexation methods, as well as physical methods, have been developed. Currently, transfection via cationic lipids or polymers is the most effective method for the delivery of negatively charged oligonucleotides and their derivatives, and many reagents are available for this approach. Such reagents are mainly cationic lipid- or cationic polymer-based compounds exploiting electrostatic interactions with negatively charged oligonucleotides to form transfection complexes. While these types of transfection reagents are quite efficient for cargo delivery to most cells in in vitro culture, their in vivo applications are rather limited largely because of general toxicity. Consequently, highly efficient, safe and robust delivery methods and agents for in vivo use are still very much in demand. New strategies have recently been devised for the delivery of biologically active molecules through the discovery of cell-penetrating peptides (CPPs) (2). CPPs are derived from a variety of different sources (e.g., venom, toxin, or synthetic) but most share common properties such as small size (less than 30 amino acid residues) and an overall cationic charge. They are generally capable of translocating across the cellular membrane through endocytotic pathway(s), and all are able to carry a variety of molecular cargoes into the cell via chemical conjugation or complexation. This method could be particularly useful for noncharged RNA interference agents such as PNAs, which – in contrast to anionic oligonucleotides – do not interact electrostatically with CPPs (3–5). To date, a variety of CPPs have been tested to improve the cellular uptake of RNA interference agents by chemical conjugation, and practically all are able to improve cellular uptake to some (greatly) varying extent without interfering with the agent’s biological (antisense) activity. However, as most CPPs utilize endocytotic pathways as their main route for cellular internalization (6–8), their bioavailability is limited because of endosomal entrapment. Although this limitation may be partly solved by the use of auxiliary reagents/treatments (e.g., chloroquine, Ca2+-treatment, photochemical internalization, etc.), which can aid by opening endo(lyso)somes to release their contents (9–11), research efforts to develop unaided delivery methods are still warranted. We have recently shown that chemical conjugation of a lipidic domain (e.g., a fatty acid such as decanoic acid) to a CPP–PNA construct significantly improves its cellular delivery (bioavailability) (12). For the evaluation of the cellular (nuclear) uptake of CPP–PNA conjugates, a splice-correction reporter system based on the use of the

Improved Cellular Uptake of Antisense Peptide Nucleic Acids

211

HeLa pLuc 705 cell line (13) was employed. HeLa pLuc 705 cells harbor an integrated, mutated luciferase gene that contains an insertion of intron 2 from human b-globin pre-mRNA carrying a cryptic splice site. This cell-based assay system allows one to gain a quantitative estimation of the cellular (nuclear) delivery efficiency of an antisense agent by measuring the level of luciferase activity that results from the correction of aberrant splicing upon targeting the agent to the cryptic splice site. This method has been widely used for studies of CPP delivery efficiency since it displays a low background level and has a positive readout with a wide dynamic range; moreover, the method is particularly suitable for stericblocking agents such as morpholino oligonucleotides (14) and PNAs since these do not induce RNase H activity.

2. Materials 2.1. Cell Culture

1. Cell growth medium: RPMI 1640 medium supplemented with 10% (v/v) fetal bovine serum (FBS) and 1% (v/v) GlutaMAX™ (Gibco). 2. HeLa pLuc 705 cell line. 3. Cell culture flask (T-25). 4. 24-well tissue culture plate. 5. Teflon cell scrapers.

2.2. Preparation of the MBHA Resin for PNA Synthesis by Downloading

1. 4-Mehylbenzhydryl amine (MBHA) resin (<1.0  mmol/g loading). 2. Glass reactor: 30-ml glass reactor fitted with a sintered glass filter. 3. Dichloromethane (DCM). 4. N,N-Dimethylformamide (DMF). 5. DMF/DCM solution (1:1, v/v). 6. Boc-protected PNA monomers (e.g., Boc-A(Z)-OH, BocC(Z)-OH, Boc-G(Z)-OH and Boc-T-OH). 7. Boc-protected amino acids. 8. N,N-diisopropylethylamine (DIPEA). 9. Activation solution: 0.057 M 2-(1-H-Benzotriazol-a-yl)1,1,3, 3-tetramethyluronium hexafluorophosphate (HBTU) in DMF. 10. Capping solution: Acetic anhydride (Ac2O)/collidine/DMF (1:1:8, v/v). 11. Kaiser test solutions: Reagent A: Dissolve 1 g of ninhydrin in 10 ml of ethanol; Reagent B: Dissolve 80 g of phenol in 20 ml

212

Shiraishi and Nielsen

of ethanol; Reagent C: Dilute 2 ml of 1 mM KCN (in water) into 100 ml of pyridine. 12. 5% (v/v) piperidine in DMF. 2.3. Synthesis of CatLip–PNA Conjugates

1. Downloaded MBHA resin obtained from Subheading  2.2, with loading downloaded to 0.12 mmol/g. 2. 4 ml glass reactor vessel. 3. DCM. 4. DMF. 5. Trifluoroacetic acid (TFA). 6. Deprotection solution for Boc synthesis reactions: Anisole/ TFA (5:95, v/v). 7. DMF/DCM (1:1, v/v). 8. Pyridine. 9. Boc-protected PNA monomers (same as in Subheading 2.2). 10. Boc-protected amino acids (same as in Subheading 2.2). 11. N,N-Diethylcyclohexylamine (DECA). 12. Activation solution: 0.085 M HBTU in DMF. 13. Kaiser test solutions (same as in Subheading 2.2). 14. Capping solution: Ac2O/collidine/DMF (1:1:8, v/v). 15. 5% (v/v) piperidine in DMF. 16. Fmoc-protected Lys monomer. 17. Fatty acids (e.g., decanoic acid). 18. Deprotection solution for Fmoc group: Piperidine/DMF (20:80, v/v). 19. Cleavage solution: TFA/trifluoromethanesulfonic (TFMSA)/m-cresol/thioanisole (6:2:1:1, v/v).

acid

20. Diethyl ether. 21. Solution for PNA solubilization: 0.1% (v/v) TFA in water. 22. High-performance liquid chromatography (HPLC) system equipped with a reversed-phase column. 23. HPLC Elution Buffer A: 0.1% (v/v) TFA in water. 24. HPCL Elution Buffer B: 0.1% (v/v) TFA in 10% (v/v) water (in acetonitrile). 25. Mass spectrometer. 2.4. Luciferase Assay of Splicing Correction by PNA Conjugates

1. Opti-MEM® medium (Invitrogen, Carlsbad, CA). Warm to room temperature before use. 2. Opti-MEM® medium containing 100 mM chloroquine (CQ). 3. PNA conjugate solution: Prepare a 200 mM working solution in water.

Improved Cellular Uptake of Antisense Peptide Nucleic Acids

213

4. Cell growth medium: RPMI 1640 medium supplemented with 1% (v/v) GlutaMAX™ and 20% (v/v) FBS (or 10% (v/v) serum replacement). 5. Phosphate-buffered saline (PBS), pH 7.2. 6. Passive Lysis Buffer (PLB) (Promega). 7. Luciferase Assay Reagent (Promega). 8. BCA Protein Assay kit (Pierce). 9. Bovine serum albumin (BSA) protein standard solutions: 0, 0.5, 1.0, 2.5, and 5.0 mg/ml BSA in PLB. 10. Luminometer instrument. 2.5. Reverse Transcription Polymerase Chain Reaction Analysis of Splicing Correction by PNA Conjugates

1. RNeasy Mini Kit (Qiagen) for purification of total RNA from cells. 2. RNase-free water: Diethyl pyrocarbonate (DEPC)-treated Milli-Q water. 3. Luciferase primers (30  mM each): Forward Primer: 5¢-TTGATATGTGGATTTCGAGTCGTC-3¢; Reverse Primer: 5¢-TGTCAATCAGAGTGCTTTTGGCG-3¢. 4. OneStep reverse transcription polymerase chain reaction (RT-PCR) Kit (Qiagen): To prepare a RT-PCR master mix solution for ten samples (88  ml, or 110% of the required ­volume), add together the following: Water, 52.8  ml; 5× buffer solution, 22 ml; dNTP mix (containing 10 mM of each dNTP), 4.4  ml; enzyme solution, 4.4  ml; and primers (forward and reverse, 30  mM each), 2.2  ml of each primer solution.

3. Methods 3.1. Cell Culture

1. Culture the HeLa pLuc 705 cells in a T-25 flask containing 8 ml of cell growth medium at 37°C under a 5% (v/v) CO2 humidified air until a fully confluent cell monolayer is obtained (see Note 1). 2. Remove the growth medium and detach the cells with a cell scraper. 3. Add 2 ml of fresh growth medium and mix gently by pipetting (see Note 2). 4. Determine the number of viable cells using a hemocytometer and trypan blue staining. 5. Transfer the cell solution (obtained in Step 3) containing 1.8 × 106 viable cells to a test tube and add fresh growth medium to a final volume of 12.5 ml (Note: This cell number is sufficient for seeding one 24-well cell culture plate).

214

Shiraishi and Nielsen

6. Dispense 0.5  ml/well of cell suspension solution into a 24-well cell culture plate (see Note 3). 7. Incubate the cells at 37°C under a humidified 5% (v/v) CO2 air overnight before performing the cell transfection experiments (see Note 4). 3.2. Preparation of Downloaded MBHA Resin for PNA Synthesis

The MBHA resin was downloaded to 0.12 mmol/g (from an initial loading of 0.8  mmol/g) in order to avoid steric hindrance during synthesis. 1. Place 1 g of MBHA resin into a 30 ml glass reactor vessel. 2. Swell the resin with DCM overnight. 3. Wash the resin five times with DMF/DCM (see Note 5). 4. Weigh out the Boc-monomer (0.12 mmol) and dissolve it in 2 ml of DMF. 5. Add 6 eq (0.72 mmol) of DIPEA (92.8 mg = 125.2 ml) to the solution in Step 4. 6. Prepare the HBTU activation solution by dissolving 0.95 eq (0.114 mmol = 43.2 mg) of HBTU in 2 ml of DMF. 7. Add the HBTU solution (Step 6) to the Boc-monomer/ DIPEA solution (Step 5). 8. Incubate the combined solution for 2 min. 9. Add the combined solution (Step 8) to the resin (Step 3). 10. Incubate the resin on a shaker overnight. 11. Wash the resin five times with DMF/DCM (see Note 5). 12. Incubate the resin in 20 ml of capping solution. 13. Wash the resin five times with DMF/DCM (see Note 5). 14. Confirm the completion of the capping reaction by performing a qualitative Kaiser test: Place a few milligrams of the dried resin into a glass tube and add one drop each of reagent A, reagent B, and reagent C. Heat the tube to 100°C for 1 min. Blue color = positive result; Yellow color = negative result. 15. Wash the resin with 5% (v/v) piperidine (in DMF) for 5 min. 16. Wash the resin five times with DMF/DCM (see Note 5). 17. Wash the resin three times with DCM (see Note 5). 18. Dry the resin in a desiccator with a vacuum pump.

3.3. Synthesis of PNA–Peptide–Lipid Conjugates

This method was originally designed for a 50 mg resin synthesis scale, but is easily adaptable to other synthetic scales (up to at least 1 g of resin). PNA synthesis was carried out by the standard Boc strategy and peptides were typically linked to the PNA at the N-terminal through an ethylene glycol linker (e.g., 8-amino3,6-dioxaoctanoic acid) or glycine via continuous solid-phase

Improved Cellular Uptake of Antisense Peptide Nucleic Acids

215

synthesis. Fatty acids (e.g., decanoic acid) were conjugated to the e-amino group of a lysine moiety during synthesis using orthogonal F-moc protection for the Lys monomer. Following final deprotection and cleavage from the solid support, the PNA–peptide–lipid conjugate product was purified and analyzed by HPLC, and further characterized by matrix-assisted laser desorption/ionization-time of flight (MALDI-TOF) mass spectrometry. 3.3.1. Synthesis

1. Place 50 mg of MBHA resin (downloaded to 0.12 mmol/g loading as described in Subheading  3.2) into a 4-ml glass reactor vessel. 2. Swell the resin with 2 ml of DCM for 18 h. 3. Initiate the PNA synthesis procedure using standard Boc chemistry: Each synthesis cycle consists of Boc-deprotection, coupling, and capping as summarized in Table 1 (see Note 5). 4. Couple Fmoc-Lys to the PNA sequence using standard Boc chemistry with the following exception: Omit the 5% (v/v) piperidine washings after capping treatment before proceeding to the following fatty acid conjugation step.

Table 1 Synthesis cycle for a PNA oligomer with Boc chemistry Synthesis step

Reagents and volume

Time (min)

Deprotection (×2)

Anisole/TFA (5:95, v/v), 1 ml

4

Wash (×3)

DMF/DCM (1:1, v/v), 4 ml

0 (Mix and remove)

Wash (×2)

Pyridine, 2 ml

Couple (×1)

0.1 M of preactivated monomer, 4 eq = 250 ml

20

Wash (×2)

DMF

0 (Mix and remove)

Capping (×1)

Ac2O/collidine/DMF (1:1:8, v/v), 1 ml

2

Wash (×3)

DMF, 2 ml

0 (Mix and remove)

Wash (×1)

5% (v/v) Piperidine in DMF

4

Wash (×3)

DMF/DCM (1:1, v/v), 4 ml

0 (Mix and remove)

Wash (×3)

DCM, 4 ml

0 (Mix and remove)

0 (Mix and remove) a

Kaiser test 

b

0.2 M of monomers in DMF and base (4 eq DECA/monomer) were activated for 1 min by mixing with an equal volume of 0.085 M HBTU in DMF b  The results of the qualitative Kaiser test should be negative: Place a few milligrams of the dried resin into a test tube and add one drop each of reagent A, reagent B, and reagent C. Heat to 100°C for 1 min. Blue color = positive result; Yellow color = negative result a 

216

Shiraishi and Nielsen

Table 2 Synthesis cycle for a PNA oligomer with Fmoc chemistry Synthesis step

Reagents and volume

Time (min)

Deprotection (×2)

Piperidine/DMF (20:80, v/v), 4 ml

10

Wash (×10)

DMF/DCM (1:1, v/v), 4 ml

Couple (×1)

0.1 M of preactivated monomer, 4 eq = 250 ml 

20

Wash (×3)

DMF/DCM (1:1, v/v), 4 ml

0 (Mix and remove)

Capping (×1)

Ac2O/collidine/DMF (1:1:8, v/v), 4 ml

2

Wash (×7)

DMF/DCM (1:1, v/v), 4 ml

0 (Mix and remove)

0 (Mix and remove) a

Kaiser test b

0.2 M of monomers in pyridine and base (2 eq DECA/monomer) were activated for 1 min by mixing with an equal volume of 0.085 M HBTU in DMF b  The results of the qualitative Kaiser test should be negative: Place a few milligrams of the dried resin into a test tube and add one drop each of reagent A, reagent B, and reagent C. Heat to 100°C for 1 min. Blue color = positive result; Yellow color = negative result a 

5. Fatty acid conjugation using Fmoc chemistry: Conjugate a fatty acid moiety to the e-amino group of the lysine residue using Fmoc synthesis starting with 20% (v/v) piperidine (in DMF) deprotection (instead of anisole/TFA) as summarized in Table 2. 6. Continue the PNA synthesis procedure until the entire sequence is completed. 3.3.2. Cleavage/ Deprotection

1. Dry the resin in a vacuum prior to cleavage. 2. Wash the resin two times with a minimum amount of TFA. 3. Incubate the resin with 250 ml of TFA/TFMSA/m-cresol/ thioanisole (6:2:1:1, v/v) for 1 h at room temperature. 4. Wash the resin two times with a minimum amount of TFA and collect the elution solution. 5. Repeat Steps 3 and 4. 6. Add 12 ml of diethyl ether to the elution solution and mix. 7. Incubate the mixture obtained in Step 6 for 20 min on ice. 8. Centrifuge the mixture (3,000 × g) at room temperature for 10–15 min to collect the precipitated PNA. Decant and discard the supernatant. 9. Wash the PNA pellet with diethyl ether three times. After each wash, collect the PNA pellet by centrifugation as described in Step 8. 10. Dry the PNA pellet in vacuo.

Improved Cellular Uptake of Antisense Peptide Nucleic Acids

217

Fig.  1. HPLC trace of a CatLip PNA–Tat conjugate with decanoic acid as the lipidic domain (Tat-Lys(Deca)–PNA: H-GRKKRRQRRRPPQ-Lys(Deca)-Gly-CCT CTT ACC TCA GTT ACA-NH2). The decanoic acid (Deca) moiety was conjugated to the e-amine of the lysine side chain. Fraction (a) (including a major peak at 23.34 min) was collected and subjected to further analysis by analytical HPLC (see Fig. 2) and mass spectrometry (see Fig. 3).

11. Dissolve the PNA pellet in 0.1% TFA and analyze the crude/ purified product by HPLC (Step 12) and mass spectrometry (Step 13). 12. Purify and analyze the crude PNA product by reversed-phase HPLC using an analytical or semipreparative column heated to 50°C. Collect appropriate elution fractions by monitoring the UV absorbance at 260 nm. A typical gradient for elution is a linear gradient of 0–50% Buffer B (0–35 min) in Buffer A. Figure 1 presents the chromatogram trace obtained by HPLC analysis of crude PNA conjugate containing a Tat peptide sequence and decanoic acid (Tat-Lys(Deca)-PNA). Figure 2 presents the chromatogram trace obtained by HPLC analysis of purified Tat-Lys(Deca)–PNA conjugate. 13. Analyze the purified PNA product using mass spectrometry. Figure 3 presents the mass spectrum graph obtained for purified Tat-Lys(Deca)–PNA conjugate. 14. Lyophilize the purified PNA conjugate and store it at 4°C until use (see Note 6). 3.4. Luciferase Assay of Splicing Correction by PNA Conjugates

1. Dilute the PNA conjugate solution (200 mM in water) with OptiMEM® medium at the desired concentration (see Note 7). 2. Remove the growth medium from the HeLa pLuc 705 cells (obtained by performing the cell culture procedures described

218

Shiraishi and Nielsen

Fig. 2. HPLC trace of purified Tat-Lys(Deca)–PNA obtained from Fraction (a) in Fig. 1.

Fig. 3. MALDI-TOF mass spectrum of purified Tat-Lys(Deca)–PNA obtained from Fraction (b) in Fig. 2. The calculated molecular weight of the Tat-Lys(Deca)–PNA conjugate is 6805.9 (found: 6806.7).

in Subheading  3.1), and then add Opti-MEM® medium containing the PNA conjugate to the cells (see Note 8). 3. Incubate the cells for 4 h (see Note 9) at 37°C under a 5% (v/v) CO2 humidified air atmosphere. 4. Add an equal volume of fresh RPMI 1640 cell growth medium containing 1% (v/v) GlutaMAX™ and 20% (v/v) FBS to the cells without removing the PNA solution (see Note 10). 5. Incubate the cells further for 24 h (see Note 11). 6. Remove the growth medium from the cells.

Improved Cellular Uptake of Antisense Peptide Nucleic Acids

219

7. Gently wash the cells once with PBS buffer solution (see Note 12). 8. Incubate the cells on a shaker with 0.1 ml/well of passive lysis buffer (PLB) until the cells are completely lysed (see Note 13). Transfer the resulting cell lysate to a new tube (see Note 14). 9. Add a 10 ml aliquot of the cell lysate to 100 ml of the luciferase assay reagent. Measure the luciferase activity using a luminometer instrument (10 s measurement with a 2 s delay) (see Note 15). 10. Normalize the measured luciferase activity by the protein amount (mg) used for the measurement and average each data point over replicates. Measure the cellular protein concentration with a BCA protein assay kit using BSA solutions as protein concentration standards (see Note 16). 3.5. RT-PCR Analysis of Splicing Correction by PNA Conjugates

1. Mix 50 ml of the HeLa pLuc 705 cell lysate (obtained from Subheading 3.4, Step 8) with 350 ml of Buffer RLT from the RNeasy Mini Kit (see Note 17). 2. Extract total RNA from the cells using the RNeasy Mini Kit by following the manufacturer’s instructions. 3. Quantify the amount of extracted total RNA by measuring the absorbance of the RNA solution at 260 nm (see Note 18). 4. Dilute the RNA solution to 1 ng/ml total RNA with RNasefree water. 5. Prepare a RT-PCR master mix solution using the OneStep RT-PCR Kit (see Note 19). 6. Add an aliquot of the total RNA solution to the master mix solution on ice (see Note 20). 7. Place the PCR tubes containing RNA samples into the PCR instrument and run the RT-PCR reactions with the following thermocycling program: ((55°C, 35  min) × 1 cycle, (95°C, 15  min) × 1 cycle, (94°C, 0.5  min; 55°C, 0.5  min; 72°C, 0.5 min) × 26–28 cycles) (see Note 21). 8. Analyze the resulting RT-PCR reaction products by electrophoresis using a 2% agarose gel.

4. Notes 1. Split the cells within 24 h after reaching full monolayer confluency in order to keep the cells growing in a healthy condition. 2. Mix the cell solution gently but thoroughly to avoid cell clumping. Use a trypsin-EDTA treatment if it is difficult to obtain single cells.

220

Shiraishi and Nielsen

3. Typically ~0.8–1 × 105 cells are seeded in each well of a 24-well cell culture plate. 4. For transfection, the cells must be exponentially growing in a healthy condition. In addition, it is recommended that the cells reach 40–60% confluency at the time of transfection. 5. Wash the resin very carefully (and thoroughly) by using a sufficient volume of solution. Remove all of the wash solution by aspiration. 6. Reconstitute the lyophilized PNA conjugate in water. If the PNA conjugate is not soluble in water, then use either DMF or DMSO. 7. Preheat the Opti-MEM® medium to room temperature. If desired, include 100  mM chloroquine (CQ) in the OptiMEM® medium for the endosomal disruption treatment. 8. Remove the entire growth medium, as serum inhibits the cellular uptake of most CPP conjugates. Moreover, this medium replacement step should be carried out fairly quickly in order to avoid cell damage. 9. Optimize the incubation time, depending on the type of CPP conjugate used. 10. Alternatively, replace (instead of supplementing) the PNA solution with growth medium containing 10% FBS in order to reduce cellular toxicity from the CPP conjugates. 11. Optimize the incubation time, depending on the experiment. 12. Wash the cells gently in order to avoid loss of cells. This step is necessary in order to obtain an accurate measurement of cellular protein (by removing the FBS components). 13. Check for complete cell lysis using a light microscope. Perform repeated freeze–thaw cycles if there is a problem with achieving complete cell lysis. 14. This solution can be stored at −20°C for at least 1 month. 15. Preheat the sample solutions and the luciferase assay reagent to room temperature. 16. If necessary, normalize the measured luciferase activity by the protein concentration of the sample. Protein concentration can also be used for cell number normalization. 17. Add 1% (v/v) 2-mercaptoethanol in Buffer RLT in order to avoid RNA degradation. In addition, use filter tips if necessary to avoid RNase contamination. 18. Check the quality of the extracted total RNA by measuring the 260/280 nm absorbance ratio. It is advisable to re-purify the extracted RNA using the RNeasy Kit if the absorbance ratio is not between 1.6 and 2.0. Check for degradation of

Improved Cellular Uptake of Antisense Peptide Nucleic Acids

221

the extracted RNA sample by gel electrophoresis if there is a problem with the RT-PCR amplification step. 19. Prepare 110% of the calculated volume of master mix solution required to perform all the RT-PCR reactions. 20. Prepare the reaction solutions on ice, and keep the tubes on ice until insertion into the PCR thermocycler to avoid nonspecific amplification. Typically, 2  ml of the total extracted RNA solution (at 1 ng/ml RNA) is added to 8 ml of the master mix solution. Optimize the required amount of template RNA sample and the required number of PCR cycles, depending on the experiment. 21. Place the PCR tubes into the thermocycler after the sample block temperature reaches 55°C in order to avoid nonspecific amplification. References 1. Torchilin, V. P. (2006) Recent approaches to intracellular delivery of drugs and DNA and organelle targeting, Annual review of biomedical engineering 8, 343–375. 2. Mae, M., and Langel, U. (2006) Cellpenetrating peptides as vectors for peptide, protein and oligonucleotide delivery, Current opinion in pharmacology 6, 509–514. 3. Debart, F., Abes, S., Deglane, G., Moulton, H. M., Clair, P., Gait, M. J., Vasseur, J. J., and Lebleu, B. (2007) Chemical modifications to improve the cellular uptake of oligonucleotides, Current topics in medicinal chemistry 7, 727–737. 4. Gait, M. J. (2003) Peptide-mediated cellular delivery of antisense oligonucleotides and their analogues, Cell Mol Life Sci 60, 844–853. 5. Zorko, M., and Langel, U. (2005) Cellpenetrating peptides: mechanism and kinetics of cargo delivery, Adv Drug Deliv Rev 57, 529–545. 6. Fotin-Mleczek, M., Fischer, R., and Brock, R. (2005) Endocytosis and cationic cell-penetrating peptides--a merger of concepts and methods, Current pharmaceutical design 11, 3613–3628. 7. Nakase, I., Niwa, M., Takeuchi, T., Sonomura, K., Kawabata, N., Koike, Y., Takehashi, M., Tanaka, S., Ueda, K., Simpson, J. C., Jones, A. T., Sugiura, Y., and Futaki, S. (2004) Cellular uptake of arginine-rich peptides: roles for macropinocytosis and actin rearrangement, Mol Ther 10, 1011–1022. 8. El-Andaloussi, S., Johansson, H. J., Lundberg, P., and Langel, U. (2006) Induction of splice

correction by cell-penetrating peptide nucleic acids, The journal of gene medicine 8, 1262–1273. 9. Abes, S., Williams, D., Prevot, P., Thierry, A., Gait, M. J., and Lebleu, B. (2006) Endosome trapping limits the efficiency of splicing correction by PNA-oligolysine conjugates, J Control Release 110, 595–604. 10. Shiraishi, T., Pankratova, S., and Nielsen, P. E. (2005) Calcium ions effectively enhance the effect of antisense Peptide nucleic acids conjugated to cationic tat and oligoarginine peptides, Chem Biol 12, 923–929. 11. Turner, J. J., Ivanova, G. D., Verbeure, B., Williams, D., Arzumanov, A. A., Abes, S., Lebleu, B., and Gait, M. J. (2005) Cellpenetrating peptide conjugates of peptide nucleic acids (PNA) as inhibitors of HIV-1 Tat-dependent trans-activation in cells, Nucleic Acids Res 33, 6837–6849. 12. Koppelhus, U., Shiraishi, T., Zachar, V., Pankratova, S., and Nielsen, P. E. (2008) Improved cellular activity of antisense peptide nucleic acids by conjugation to a cationic peptide-lipid (CatLip) domain, Bioconjug Chem 19, 1526–1534. 13. Kang, S. H., Cho, M. J., and Kole, R. (1998) Up-regulation of luciferase gene expression with antisense oligonucleotides: implications and applications in functional assay development, Biochemistry 37, 6235–6239. 14. Summerton, J., and Weller, D. (1997) Morpholino antisense oligomers: design, pre­ paration, and properties, Antisense Nucleic Acid Drug Dev 7, 187–195.

Chapter 14 Synthesis of Oligonucleotide–Peptide Conjugates for Biomedical and Technological Applications Anna Aviñó, Santiago Grijalvo, Sónia Pérez-Rentero, Alejandra Garibotti, Montserrat Terrazas, and Ramon Eritja Abstract Oligonucleotide–peptide conjugates have attracted considerable interest especially for biomedical uses. In the first part of this chapter, we describe protocols for the stepwise synthesis of oligonucleotides carrying peptide sequences at the 3¢-end on a single support. The resulting oligonucleotide–peptide conjugates may be used as exogenous effectors for the specific control of gene expression. In the second part of this chapter, detailed postsynthetic conjugation protocols to introduce peptide sequences into oligonucleotide sequences are also presented. Key words: DNA, RNA, Nucleic acids, RNA interference, Antisense, Oligonucleotide–peptide conjugates, Stepwise synthesis, Postsynthetic conjugation

1. Introduction Oligonucleotide–peptide conjugates are chimeras of oligonucleotides and peptides that are produced in order to add some of the biological and/or biophysical properties of peptides to oligonucleotides. A peptide sequence may be attached to oligonucleotides either at the 5¢- or 3¢-terminal ends, or linked to the nucleobases at the internal positions (Fig.  1). Most frequently, peptides are linked at the terminal positions of oligonucleotides since their attachment at internal sites can potentially interfere with the basepairing properties of the nucleobases. Peptide–oligonucleotide conjugates are prepared through distinct conjugation chemistries, which include either stable or cleavable linkages. In the postsynthetic conjugation approach, the two moieties are prepared

Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_14, © Springer Science+Business Media, LLC 2011

223

224

Aviñó et al.

B

O

OH

PEPTIDE

O O−

O−

P

O R

O

O

B

O R

O

O

P

B

O

O

O

O − O P O R

O

B

O

O

n O

B

−

B

O

OH

O − O P O R O

O O− P

LINKER

O − O P O

O

P

O R

O

O

n B

LINKER

O

PEPTIDE

B

O

n

O− P

O R

O

O

B

O O

−

P O

O R LINKER at the 3'-end

HO R PEPTIDE at the 5'-end

HO R

R = H, DNA R = OH, RNA

at the nucleobase

Fig. 1. Illustration of the potential sites within an oligonucleotide that is available for the attachment of peptides.

independently using solid-phase synthesis, and then thiols and maleimido groups are specifically introduced to link the two molecules together (1). In the stepwise synthesis approach, oligonucleotide–peptide conjugates are prepared by the addition of protected amino acids and nucleotides during solid-phase synthesis on the same solid support (2–4). In this case, the main challenge is to develop an effective protection strategy for peptide synthesis that is compatible with oligonucleotide synthesis. For example, during the final steps of solid-phase peptide synthesis a treatment with acid is usually required, which can lead to partial depurination of DNA oligonucleotides. In the case of the synthesis of oligonucleotide 3¢-peptide conjugates, the solution to this issue is the use of tert-butoxycarbonyl (Boc)-protected amino acids carrying fluorenylmethyl (Fm) or fluorenylmethoxycarbonyl (Fmoc) groups for the protection of the side chains (2–4). This unusual Boc/Fmoc strategy, described in several recent reviews (5–9), is compatible with oligonucleotide synthesis, and most of the required amino acid derivatives can be obtained from commercial sources. Herein, we describe protocols for the preparation of oligonucleotide–peptide conjugates developed in our research group at IRB. Specifically, we describe the stepwise synthesis of DNA and RNA molecules covalently linked to peptides that are designed to enhance cellular uptake in antisense (10) and RNA interference (11, 12) experiments. In addition, the synthesis of DNA molecules carrying peptide epitopes by postsynthetic conjugation (13) is also discussed.

Synthesis of Oligonucleotide–Peptide Conjugates for Biomedical

225

2. Materials 2.1. Solid-Phase Synthesis of the Peptide Moiety

1. Boc-amino acids carrying base-labile groups are standard reagents that are commercially available from a number of suppliers. The protecting groups of lysine and ornithine are either Fmoc or trifluoroacetyl. Glutamic and aspartic acid are protected with the Fm ester. Threonine, tyrosine, and serine are protected with the acetyl group. Histidine can be protected with the tosyl group, and cysteine with the Fm or the t-butylthio group (7). The remaining amino acids were used without protection. Arginine can be protected with the di-Fmoc protecting group, but this derivative is not commercially available at present (see Note 1). 2. The coupling agent for peptide synthesis was benzotriazol-1yl-oxy-tris(pyrrolidino)phosphonium hexafluorophosphate (PyBOP). 3. The solid-phase synthesis support, amino-polyethylene glycol-polystyrene (PEG-PS), was obtained from PerSeptive Biosystems.

2.2. Synthesis of Oligonucleotides Carrying Peptide Sequences at the 3 ¢-End

1. Oligonucleotide sequences were prepared by solid-phase synthesis using 2-cyanoethyl phosphoramidites as monomers. The phosphoramidites of the natural nucleosides are standard reagents that are commercially available (see Note 2). The base protection scheme employed during DNA synthesis was as follows: 2¢-deoxyguanosine was protected with either the isobutyryl group or the dimethylaminomethylidene group; and 2¢-deoxycytidine and 2¢-deoxyadenosine were protected with the benzoyl group. Base protection during RNA synthesis was as follows: guanosine was protected with the dimethylaminomethylidene group; cytidine was protected with the acetyl group; and adenosine with the benzoyl group. The 2¢OH protecting group employed for the RNA monomers was the t-butyldimethylsilyl (TBDMS) group. 2. Ancillary reagents used during oligonucleotide synthesis included the following: 0.4  M 1H-tetrazole in acetonitrile (ACN) (activation); 3% trichloroacetic acid in dichloromethane (DCM) (detritylation), acetic anhydride/pyridine/tetrahydrofuran (1:1:8, v/v/v) (capping A), 10% (v/v) N-methylimidazole in tetrahydrofuran (capping B), 0.01 M iodine in tetrahydrofuran/pyridine/water (7:2:1, v/v/v) (oxidation). All of the preceding solutions can be obtained from the same companies that provide phosphoramidites (see Note 2). 3. 6-Aminohexanol.

226

Aviñó et al.

4. Succinic anhydride. 5. N,N-Dimethylaminopyridine (DMAP). 6. 4-Hydroxybutyric acid. 7. Preparative high-performance liquid chromatography (HPLC) system equipped with a Nucleosil® 120-10 C18 column (250 × 4 mm) (Macherey-Nagel, Inc.). 8. Analytical HPLC system equipped with an XBridgeTM OST C18 column (2.5 mm, 4.6 × 50 mm) (Waters). 9. Solvent A for HPLC: 5% ACN in 100 mM aqueous triethylammonium acetate (TEAA), pH 6.5. 10. Solvent B for HPLC: 70% ACN in 100 mM aqueous TEAA, pH 6.5. 11. Voyager-DETMRP mass spectrometer equipped with a nitrogen laser (337 nm) (PerSeptive Biosystems). 12. Polyacrylamide gel electrophoresis system (Hoefer Scientific). 13. Stains-All nucleic acid and protein staining solution (Sigma– Aldrich) was prepared at 0.01% (w/v) in formamide/water (1/1.2, v/v). 2.3. Postsynthetic Conjugation of Peptides to Oligonucleotides Carrying Thiol Groups

1. The following reagents were used for the introduction of thiol groups into oligonucleotides and were obtained from commercial sources. At the 3¢-end: 3¢-thiol-modifier C3 S-S CPG and at the 5¢-end: 5¢-thiol modifier C6 S-S phosphoramidite (see Note 2 and Fig. 2). A 2¢-deoxycytidine derivative was used for the introduction of a thiol group at internal nucleobase positions (Fig. 2), as described in ref. 13. O S

DMT-O

S

NH-CPG

O

at the 3' end

O DMT O

S

O

S S

NH

O P CN N(C3H7)2

at the 5' end

S

N DMT O

O

O

P

N

O at the nucleobase

O

CN

N(C3H7)2

Fig. 2. Reagents used for the synthesis of modified oligonucleotides containing thiol groups.

Synthesis of Oligonucleotide–Peptide Conjugates for Biomedical

227

2. For the introduction of maleimido groups at the N-terminal position of peptides, 3-maleimidobenzoic acid N-hydroxysuccinimide ester (MBS) was used. 3. Rink-Amide polystyrene solid support was used for the synthesis of peptides by the Fmoc/t-butyl strategy. 4. Standard Fmoc-amino acids protected with t-butyl groups were used; and HATU, (O-7-azabenzotriazole-1-yl)-N,N,N ¢,N ¢tetraethyluronium hexafluorophosphate, was used as a coupling agent. 5. Illustra NAP™-10 desalting column prepacked with Sephadex® G-25 (GE Healthcare).

3. Methods There is currently great interest in the use of oligonucleotide–peptide conjugates for a variety of biomedical applications. As dis­cussed in the introduction to this chapter, there are two major approaches that may be used to prepare these types of compounds (see Note 3). In the postsynthetic conjugation approach, the oligonucleotide and peptide components are built on separate supports using standard solid-phase protocols but are conveniently functionalized so that they may be linked together after synthesis. In the stepwise solid-phase approach, the oligonucleotide–peptide conjugate is prepared on a single support using special protecting groups in conjunction with modified synthesis protocols that minimize unwanted side-reactions. In this section, we describe specific protocols that our group has developed in recent years; several alternative protocols reported by others may also be found in the bibliography (5–9). As an example of the scope of the utility of the approaches described in this chapter, Table 1 lists several oligonucleotide–peptide conjugates that have been produced by our group using these protocols. Prior to conducting any oligonucleotide–peptide synthesis work, firstly, a decision to employ either a stepwise or a postsynthetic conjugation protocol should be made. Postsynthetic conjugation protocols tend to offer a greater level of flexibility, as peptides can be introduced either at the 3¢ or 5¢ ends of the oligonucleotides or at internal nucleobases. Moreover, if the peptides are long and complex or if the oligonucleotides are long and highly modified, then postsynthetic conjugation protocols are generally more suitable. Unfortunately, not all types of peptides and oligonucleotides work well in these protocols. For example, these conjugation reactions are not very efficient when using hydrophobic peptides that are not very soluble in aqueous solvents. In addition, low yields have been observed during the conjugation of oligonucleotides to highly structured peptides with several

Nonradioactive labeling, 2D-DNA arrays

Nonradioactive labeling

Antisense and siRNA inhibition of gene expression Binding to jun

Postsynthetic

Stepwise

Postsynthetic Stepwise

Postsynthetic

c-myc a

c-myc a

NLS SV40 b Nucleoplasminec Nucleoplasminc -(Lys)n- n = 2, 4, 8 -(Arg)n- n = 2, 4, 8 fos d

b 

a 

Antisense inhibition of gene expression

Use

Protocol

Peptide

Peptide epitope recognized by the monoclonal antibody from clone 9E10 Nuclear localization sequence of simian virus SV40 c  Nuclear localization sequence of nucleoplasmin d  Peptide fragment (35 amino acids) containing the leucine-zipper region of fos

Oligonucleotide

Table 1 Example of oligonucleotide–peptide conjugates prepared by using the protocols described in this chapter

(16)

(3, 4, 12)

(1, 4)

(14, 15)

(13)

References

228 Aviñó et al.

Synthesis of Oligonucleotide–Peptide Conjugates for Biomedical

229

positive charges. On the other hand, stepwise solid-phase protocols are efficient only when using short-length peptides and moderatesized oligonucleotides. Furthermore, some amino acids such as arginine are difficult to introduce, as there is no good protecting group available for the guanidino group that can be removed with base (see Note 1). And finally, stepwise protocols are only efficient when synthesizing conjugates with the peptide linked at the 3¢-end of the oligonucleotide (see Note 4). In the stepwise protocol presented below, oligonucleotide–peptide conjugates were synthesized on a PEG-PS support. This support was selected since it gave the best results for the coupling of amino acids and nucleoside phosphoramidites. Figure  3 outlines the general procedure for the stepwise synthesis of oligonucleotide– peptide conjugates on the same support. To avoid the use of strong acids in the presence of the oligonucleotide, the peptide component was synthesized first using the acid-labile Boc group to protect the a-amino function. The protecting groups for the

3.1. Preparation of Oligonucleotide– Peptide Conjugates by Stepwise SolidPhase Synthesis

a

O NH2

O

NH

(CH2)6

NH

Boc

O

b

O O

NH

(CH2)6

NH

O

c

Boc

PEPTIDE

NHCO(CH2)3O-Trityl

O

O O

NH

PEPTIDE

(CH2)6

NH

O

O

OH

d

B

O

O O

−

P

R

O

O

B

O O

O

−

P

n

R

O

O

B

O O

O HO

(CH2)6

PEPTIDE

NH

−

P

O

R

NHCO(CH2)3O

O

Fig. 3. Outline of the preparation of oligonucleotide-3¢-peptide conjugates using stepwise solid-phase synthesis. (a) Boc–NH–(CH2)6–OCO–(CH2)2–COOH, PyBOP; (b) peptide synthesis using Boc-amino acids, PyBOP; (c) trityl–O–(CH2)3– COOH, PyBOP; (d) oligonucleotide synthesis, ammonia deprotection, and HPLC purification.

230

Aviñó et al.

side chains of the amino acids (Fmoc for Lys and Fm for Asp) and the linker for the first amino acid to the support were base-labile so that they could be removed at the same time as the protective groups for the nucleobases. To this end, 6-aminohexylsuccinyl was utilized as the linker molecule, yielding a peptide with an aminohexylamide group at the C-terminal. Once the peptide had been synthesized, the spacer molecule 4-O-trityl 4-hydroxybutyric acid was incorporated to allow the peptide to connect with the oligonucleotide component. This spacer molecule introduces a hydroxyl group protected by a trityl group at the N-terminal position of the peptide. After the addition of the linker, the oligonucleotide sequences were assembled. At the end of the synthesis, the oligonucleotide–peptide conjugates were purified by HPLC and characterized by matrix-assisted laser desorption/ionization-time of flight (MALDI-TOF) mass spectrometry (MS). Note that this protocol may be used for the synthesis of either DNA or RNA sequences for the oligonucleotide component; in the latter case, the RNA monomers are protected with a TBDMS group. Furthermore, oligodeoxynucleotides carrying lysine or ornithine residues can have these converted to homoarginine or arginine by reaction with O-methylisourea. 3.1.1. Solid-Phase Synthesis of the Peptide Moiety

1. Boc-6-aminohexyl hemisuccinate was synthesized as follows: Boc-aminohexanol (1 eq) was reacted with succinic anhydride (1.5 eq) and DMAP (1.5 eq) in DCM at room temperature overnight to form the corresponding hemisuccinate. The mixture was next dissolved with additional DCM and treated with a solution of 0.1 M NaH2PO4 (pH 5). The organic layer was dried with anhydrous sodium sulfate and concentrated to dryness. The resulting compound was used in the next step without further purification. 2. PEG-PS support was washed with DCM (4 × 30  s); trifluoroacetic acid (TFA)/DCM (2:3, v/v; 1 × 1 min; 1 × 30 min); DCM (4 × 30 s); N,N-diisopropylethylamine (DIEA)/DCM (1:19, v/v; 3 × 1 min) and DCM (4 × 30 s). 3. Boc-6-aminohexyl hemisuccinate (synthesized in step 1) was reacted with the solid support for 2 h at room temperature using the following mixture: PyBOP (5.0  eq), DIEA (10.0 eq), and hemisuccinate (5.0 eq). This mixture was previously preactivated (5  min) in N,N-dimethylformamide (DMF) and introduced later onto the solid support. After the coupling of the hemisuccinate, the resulting support was acetylated using a mixture of acetic anhydride/DIEA/DMF (1:1.7:15.3, v/v/v; 15 min). 4. The peptide chain was elongated manually in DMF using a fivefold molar excess of amino acid Boc-protected amino acids and a fivefold molar excess of PyBOP and a tenfold

Synthesis of Oligonucleotide–Peptide Conjugates for Biomedical

231

molar excess of DIEA for 1 h. The Boc group was removed using the following procedure: solid-phase support was treated with DCM (4 × 30 s); TFA/DCM (2:3, v/v; 1 × 1 min; 1 × 30  min); DCM (4 × 30  s); DIEA/DCM (1:19, v/v; 3 × 1 min); and DCM (4 × 30 s). Boc-amino acids were protected with base-labile groups. 5. The trityl derivative of 4-hydroxybutyric acid was synthesized as follows (see Note 5): 1  eq of 4-hydroxybutyric acid was reacted with 1.4  eq of triphenylchloromethane (trityl chloride) in anhydrous pyridine overnight at 50°C. The reaction was stopped by the addition of methanol, and the solvent was evaporated. Silica gel chromatography using DCM with an increasing amount of methanol (0–5%) plus 1% triethylamine yielded the desired product at 50% yield. 6. Coupling of the trityl derivative of 4-hydroxybutyric acid to the peptide: The addition of this building block introduces a hydroxyl group to the existing peptide. This hydroxyl group is required for the subsequent assembly of the oligonucleotide (see Subheading 3.1.2). The linker was coupled to the peptide-support using PyBOP and DIEA as described above in step 3. Afterward, any remaining unreacted amino groups on the support were blocked with acetic anhydride and DIEA. 3.1.2. Synthesis of Oligonucleotides Carrying Peptide Sequences at the 3¢-Ends

1. The peptide-modified supports prepared, as described in Subheading  3.1.1, were next employed for the synthesis of the oligonucleotide sequences in a DNA synthesizer using the following procedure: 2-cyanoethyl phosphoramidites were dissolved (to give 0.1 M solutions) in either dry DCM, or ACN, or with a mixture of ACN/DCM (8:2, v/v) (see Note 6). A modified synthesis cycle was used in which the coupling time was increased to 5 min, the capping and oxidation times to 1  min, and the detritylation time to 2  min (4 × 30 s). In the DNA sequences, the last 4,4¢-dimethoxytrityl (DMT)-protecting group was not removed. In the RNA sequences, the last DMT group was removed because the DMT is partially lost during the fluoride treatment (see Note 7). The average coupling yield was ~97–98% per step for RNA monomers and 99% for 2¢-O-methyl-RNA and DNA monomers. 2. The solid supports containing the oligonucleotide–peptide conjugates were washed with ACN, treated with a solution containing 0.5  M 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU) in ACN for 5 min, washed with ACN, and then dried (see Note 8). In order to deprotect conjugates containing DNA oligonucleotides, step 3 below was next carried out; for conjugates containing RNA oligonucleotides, step 4 was carried out instead.

232

Aviñó et al.

3. For the deprotection of DNA-peptide conjugates: The solid supports were treated with concentrated aqueous ammonia overnight at 55°C and then recovered by filtration. After filtration, the supports were washed with water and the combined solutions were evaporated to dryness. Following this, HPLC purification of the DNA-peptide conjugates was carried out by proceeding directly to step 5 below. 4. For the deprotection of RNA-peptide conjugates: The solid supports were treated with concentrated aqueous ammonia– ethanol (3:1, v/v) for 1  h at 55°C and then recovered by filtration. After filtration, the supports were washed with ethanol and the combined solutions were evaporated to dryness. The supports were next treated with 0.15 ml of triethylamine tris(hydrofluoride)/triethylamine/N-methylpyrrolidone (4:3:6, v/v/v) at 65°C to remove the TBDMS groups. After incubating for 2.5 h at 65°C, the reactions were stopped by the addition of 0.3  ml of isopropoxytrimethylsilane and 0.75 ml of diethyl ether. The resulting mixtures were stirred and cooled down to 4°C, which led to the formation of a precipitate. The precipitate material was collected by centrifugation at 4,500 × g for 5 min at 4°C, washed with ether and then centrifuged again. Finally, the collected residues were dissolved in water and the RNA–peptide conjugates were purified by HPLC as described in step 5 below. 5. Reversed-phase HPLC purification of oligonucleotide– peptide conjugates was performed using a Nucleosil® 120-10 C18 column (250 × 4 mm) under the following conditions for solvent A and solvent B: 20 min linear gradient from 15 to 80% solvent B, and 5 min 80% solvent B (DMT-on conditions); 20 min linear gradient from 0 to 50% B (DMT-off conditions); flow rate of 3 ml/min. 6. Characterization of oligonucleotide–peptide conjugates by mass spectrometry (MS) was performed as follows: MALDITOF MS spectra were acquired with a mass spectrometer equipped with a nitrogen laser at 337 nm using a 3-ns pulse. The matrix contained 2,4,6-trihydroxyacetophenone (THAP; 10 mg/ml in ACN/water, 1:1, v/v) and ammonium citrate (50 mg/ml in water) at a ratio of 9:1 (v/v). 7. Analytical reversed-phase HPLC was performed using a Waters XBridge™ OST C18 column (2.5  mm, 4.6 × 50  mm) under the following conditions for solvent A and solvent B: 10 min linear gradient from 0 to 35% B; flow rate of 1 ml/min (Fig. 4). 8. Characterization of oligonucleotide–peptide conjugates by denaturing polyacrylamide gel electrophoresis was performed as follows: Polyacrylamide gels (20%, w/v; 8 M urea; acrylamide/bisacrylamide ratio = 19:1, w/w; 14 × 16 × 0.1 cm) were

Synthesis of Oligonucleotide–Peptide Conjugates for Biomedical

233

Fig.  4. A representative high-performance liquid chromatography (HPLC) purification ­profile for an oligonucleotide-3¢-peptide conjugate product obtained by stepwise synthesis.

run for 5 h at 400 V. Three micrograms of oligonucleotide– peptide conjugates were loaded in each lane of the gel. Following electrophoresis, gels were stained with Stains-All solution to visualize the oligonucleotide–peptide conjugate bands. 3.1.3. Introduction of Guanidine Groups

Oligodeoxynucleotide–peptide conjugates carrying lysine or ornithine residues can have these be converted to homoarginine or arginine by reaction with O-methylisourea as follows: 1. Oligodeoxynucleotide–peptide conjugates containing unprotected amines (~4.0 OD) were treated with a mixture (125 ml) prepared by combining 100 ml of a solution containing 1 g/ml O-methylisourea chloride in water with aqueous ammonia (32%, w/w; 125 ml). 2. After heating the above reaction mixture at 55°C overnight, the solvent was removed. 3. The resulting modified conjugates were desalted by Sephadex® gel filtration, and further purified by preparative HPLC (see Subheading 3.2, step 5).

3.2. Preparation of Oligonucleotide– Peptides Conjugates by Postsynthetic Conjugation of Peptides with Oligonucleotides Carrying Thiol Groups

1. Firstly, a peptide carrying maleimido groups was synthesized using a Rink-Amide polystyrene support (0.56 mmol/g loading) and Fmoc-amino acids. The Fmoc protecting groups were removed by treating the resin with a 1:1 (v/v) solution of piperidine/DMF for 2 min, and then for 8 min with a solution of 20% (v/v) piperidine in DMF. Following this, the support was washed four times with DMF. Next, a mixture containing 4  eq of Fmoc-protected amino acid (0.9  mmol), 4  eq of N-hydroxybenzotriazole (HOBt) (0.121 g, 0.9 mmol), 4 eq of

234

Aviñó et al.

2-(1H-benzotriazol-1-yl)-1,1,3,3-tetramethy-luronium hexafluorophosphate (HBTU) (0.34 g, 0.9 mmol), 8 eq of DIEA (312 ml, 1.8 mmol), and a small amount of DMF was preactivated for 5  min before the addition to the resin. The completion of the coupling reaction was qualitatively checked with the ninhydrin test. To block unreacted amino groups, a mixture of acetic anhydride (1 ml), DIEA (1.7 ml) and DMF (15.3 ml) was added for 15 min. 2. After completion of the peptide sequence, 6 eq of MBS and 6 eq of HOBt in a small amount of DMF were added. The maleimido-peptide was deprotected and released from the resin by treatment with TFA containing 5% (v/v) water for 4 h at room temperature. Following this, the resin was filtered and washed several times with water. The combined solutions were evaporated under reduced pressure. The resulting residue was washed three times in diethyl ether and concentrated to dryness to eliminate the excess TFA. 3. Oligonucleotides carrying a thiol group at either the 3¢-end, the 5¢-end or at an internal nucleobase position were synthesized on a DNA synthesizer using the standard ß-cyanoethylprotected phosphoramidite method by following protocols recommended by the supplier (Fig.  5). The oligonucleotides were removed from the solid support by treatment with concentrated Succ

OLIGONUCLEOTIDE

S-S-R

OLIGONUCLEOTIDE

SH O

PEPTIDE

N O

OLIGONUCLEOTIDE

S

O N

PEPTIDE

O

Fig. 5. Outline of the preparation of oligonucleotide–peptide conjugates using the postsynthetic conjugation method. An oligonucleotide carrying a thiol group either at the 3¢- or 5¢-ends, or at an internal nucleobase is assembled using conventional solid-phase synthesis protocols. The use of a 0.1-M dithiothreitol solution during ammonia deprotection removes all the protecting groups and also cleaves the disulfide bond, yielding oligonucleotides with a free thiol group that can react with maleimido-modified peptides.

Synthesis of Oligonucleotide–Peptide Conjugates for Biomedical

235

ammonia containing 0.1  M dithiothreitol (DTT) for 6  h at 55°C (see step 4 below). Under these conditions, all base-protecting groups were also cleaved. 4. To prevent side reactions, the oligonucleotides and maleimido-modified peptides were cleaved from the solid support just before use. For the same reason, the maleimido-peptide was used without further purification. After synthesis, the oligonucleotide was treated with a solution (1  ml) of 0.1  M DTT in concentrated ammonia at 55°C overnight. Under these conditions, the tert-butyl thiol-protecting group was removed. The solution was next concentrated to dryness. Following this, the residue thus obtained was dissolved in sterile water. DTT was removed from the crude oligonucleotide solution by using a NAP™-10 desalting column. The oligonucleotide was eluted with 1.5 ml sterile water. The purified oligonucleotide solution was then mixed with 0.5 ml of solvent A containing a tenfold molar excess of the maleimidopeptide (Fig. 5). To adjust the solution to pH 6.5, a few drops of 1  M aqueous TEAA solution (pH 6.5) were added (see Note 9). After 10 h at room temperature, the reaction mixture was concentrated to dryness, dissolved in sterile water, and purified with HPLC (Fig. 6). The final oligonucleotide– peptide conjugate was characterized by mass spectrometry. p18 +tiol t8 0,14

10,98049

Abs 260 nm

0,12 0,10 0,08

8,99173

0,06 0,04 0,02 0,00

−0,02

10,01695 0

5

10

15

20

Tr(min)

Fig.  6. A representative HPLC purification profile for the conjugation products obtained by reacting a N-maleimidopeptide (15 amino acids) with a 5¢-thiol-oligonucleotide (eight bases). The desired oligonucleotide–peptide conjugate appears as a double peak (two isomers) eluting at approximately 11 min. The peak eluting at ~9 min is the starting thiol-modified oligonucleotide. For conjugates that contain longer oligonucleotides, the separation of the two isomers cannot typically be observed.

236

Aviñó et al.

4. Notes 1. The protection of arginine is difficult, as there is no good protecting group available for the guanidino group that can be removed in the presence of oligonucleotides. The use of a Di-Fmoc derivative of arginine has been reported by others (17); here, we describe a protocol for the guanidinylation of ornithine-peptides that yields arginine-peptides in excellent yields. This protocol can also be used for the synthesis of homoarginine peptides from lysine peptides. 2. There are several companies, such as Applied Biosystems and Sigma–Aldrich that supply conventional phosphoramidites, ancillary reagents, and solid supports for oligonucleotide synthesis. In addition to the standard reagents, more elaborated phosphoramidites and solid supports can be obtained from companies that specialize in oligonucleotide reagents, such as Link technologies (Scotland), Glen Research (Sterling, VI), Berry & Associates (Dexter, MI), TriLink Biotechnologies (San Diego, CA), Nedken (Foster City, CA), ChemGenes (Wilmington, MA), Prime Synthesis (Aston, PA, USA), and Metkinen (Finland). 3. Additionally, it is possible to use a third method, known as solid-phase fragment condensation (SPFC), in which a partially protected peptide fragment is coupled directly to an oligonucleotide on a solid support (10). 4. Protocols for the assembly of peptides at the 5¢-end of oligonucleotides have been described previously (18, 19). In general, peptide assembly at the 5¢-end of oligonucleotides is less efficient due to side reactions such as acetylation. 5. 6-Hydroxyhexanoic acid and 12-hydroxydodecanoic acid derivatives can be prepared in a similar manner. 6. We have observed that the specific peptide sequence present on the solid-phase support can influence coupling yields during oligonucleotide synthesis. Supports carrying a large number of lysines protected with Fmoc groups show less efficient coupling when phosphoramidites are dissolved in ACN. In this case, coupling efficiencies are improved if the phosphoramidites are dissolved in DCM. Depending on the peptide sequence, ACN, DCM, or mixtures of both solvents may be used. 7. The use of tetrabutylammonium fluoride solution instead of triethylamine tris(hydrofluoride) does not remove the last DMT; however, removal of the silyl groups is slower. 8. The DBU treatment eliminates the cyanoethyl groups from phosphates. This step is important to avoid the addition of acrylonitrile to lysine residues.

Synthesis of Oligonucleotide–Peptide Conjugates for Biomedical

237

9. During the conjugation of 5¢-thiol oligonucleotides with a peptide carrying the leucine-zipper region of fos, we have observed low yields using TEAA buffers. However, using 10  mM Tris–HCl, 100  mM NaCl, pH 7.5 as the buffer improved the yields dramatically.

Acknowledgments This work is supported by the Spanish Ministry of Education (BFU2007-63287/BMC), Generalitat de Catalunya (2009/ SGR/238), Instituto de Salud Carlos III (CIBER-BNN), and European Communities (DYNAMO contract 028669 (NEST), FUNMOL NMP4-SL-2009-213382). References 1. Eritja, R., Pons, A., Escarceller, M., Giralt, E., and Albericio, F. (1991) Synthesis of defined peptide-oligonucleotide hybrids containing a nuclear transport signal sequence. Tetrahedron 47, 4113–4120. 2. Soukchareun, S., Tregear, G.W., and Haralambidis, J. (1995) Preparation and characterization of antisense oligonucleotide-­ peptide hybrids containing viral fusion peptides. Bioconjug. Chem. 6, 43–53. 3. de la Torre, B.G., Aviñó, A., Tarrasón, G., Piulats, J., Albericio, F., and Eritja, R. (1994) Stepwise solid-phase synthesis of oligonucleotidepeptide hybrids. Tetrahedron Lett. 35, 2733–2736. 4. de la Torre, B.G., Albericio, F., SaisonBehmoaras, E., Bachi, A., Eritja, R. (1999) Synthesis and binding properties of oligonucleotides carrying nuclear localization sequences. Bioconjug. Chem. 10, 1005–1012. 5. Tung, C.H., and Stein, S. (2000) Preparation and applications of peptide-oligonucleotide conjugates. Bioconjug. Chem. 11, 605–618. 6. Eritja, R. (2000) Synthesis of oligonucleotidepeptide conjugates and nucleopeptides. SolidPhase Synthesis. A practical guide. (S. Kates, F. Albericio Eds) Marcel Dekker, New York, pp529-548. 7. Grandas, A., Marchán, V., Debéthune, L., and Pedroso, E. (2004) Stepwise solid-phase synthesis of nucleopeptides. Current Protocols in Nucleic Acid Chemistry, John Willey & Sons, NY, Chapt. 4.22.1–4.22.54. 8. Röglin, L., Seitz, O. (2008) Controlling the activity of peptides and proteins with smart

nucleic acid-protein hybrids. Org. Biomol. Chem. 6, 3881–3887. 9. Venkatesan, N., and Kim, B.H. (2006) Peptide conjugates of oligonucleotides: synthesis and applications. Chem. Rev. 106, 3712–3761. 10. Gait, M.J. (2003) Peptide-mediated cellular delivery of antisense oligonucleotides and their analogues. Cell Mol. Life Sci. 60, 844–853. 11. Brumcot, D., Manoharan, M., Koteliansky, V., and Sah, D.W. (2006) RNAi therapeutics: a potential new class of pharmaceutical drugs. Nat. Chem. Biol. 2, 711–719. 12. Aviñó, A., Ocampo, S. M., Caminal, C., Perales, J.C., and Eritja, R. (2009) Stepwise synthesis of RNA conjugates carrying peptide sequences for RNA interference studies. Mol. Divers. 13, 287–293. 13. Gottschling, D., Seliger, H., Tarrasón, G., Piulats, J., Eritja, R (1998) Synthesis of oligodeoxynucleotides containing N4mercaptoethylcytosine and their use in the preparation of oligonucleotide-peptide conjugates carrying c-myc tag sequence. Bioconjug. Chem. 9, 831–837. 14. Frieden, M., Aviñó, A., Tarrasón, G., Escorihuela, M., Piulats, J., Eritja, R. (2004) Synthesis of oligonucleotide-peptide conjugates carrying the c-myc peptide epitope as recognition system. Chem. Biodivers. 1, 930–938. 15. Merkoçi, A., Aldavert, M., Tarrasón, G., Eritja, R., Alegret, S. (2005) Toward an ICPMS-Linked DNA assay based on gold nanoparticles immunoconnected through peptide sequences. Anal. Chem. 77, 6500–6503.

238

Aviñó et al.

16. Portela, C., Albericio, F., Eritja, R., Castedo, L., Mascareñas, J.L. (2007) Ds-Oligonu-cleotidepeptide conjugates featuring peptides from the leucine zipper region of fos as switchable receptors for the oncoprotein Jun. ChemBioChem 8, 1110–1114. 17. Debethune, L., Marchan, V., Fabregas, G., Pedroso, E. Grandas, A. (2002) Towards nucleopeptides containing any trifunctional amino acid (II) Tetrahedron 58, 6955–6978.

18. Zaramella, S.; Yeheskiely, E.; Strömberg, R. (2004) A method for solid-phase synthesis of oligonucleotide 5’-peptide-conjugates using acid-labile alpha-amino protections. J. Am. Chem. Soc. 126, 14029–14035. 19. Ocampo, S. M., Albericio, F., Fernández, I., Vilaseca, M., Eritja, R. (2005) A straightforward synthesis of 5’-peptide oligonucleotide conjugates using Na-Fmoc-protected amino acids. Org. Lett. 7, 4349–4352.

Chapter 15 Amphiphilic DNA Block Copolymers: Nucleic Acid-Polymer Hybrid Materials for Diagnostics and Biomedicine Jan Zimmermann, Minseok Kwak, Andrew J. Musser, and Andreas Herrmann Abstract DNA-polymer conjugates have been recognized as versatile functional materials in many different fields ranging from nanotechnology to diagnostics and biomedicine. They combine the favorable properties of nucleic acids and synthetic polymers. Moreover, joining both structures with covalent bonds to form bioorganic hybrids allows for the tuning of specific properties or even the possibility of evolving completely new functions. One important class of this type of material is amphiphilic DNA block copolymers, which, due to microphase separation, can spontaneously adopt nanosized micelle morphologies with a hydrophobic core and a DNA corona. These DNA nano-objects have been explored as vehicles for targeted gene and drug delivery, and also as programmable nanoreactors for organic reactions. Key to the successful realization of these potential applications is that (1) DNA block copolymer conjugates can be fabricated in a fully automated fashion by employing a DNA synthesizer; (2) hydrophobic compounds can be loaded within their interior; and (3) they can be site-specifically functionalized by a convenient nucleic acid hybridization procedure. This chapter aims to broaden the range of biodiagnostic and biomedical applications of these materials by providing a comprehensive outline of the preparation and characterization of multifunctional DNA-polymer nanoparticles. Key words: DNA, Block copolymers, Nanoparticles, Oligonucleotides, Drug delivery, Anticancer drug, Macromolecular amphiphiles, Solid-phase synthesis

1. Introduction The first reports of DNA-polymer hybrids go back to the late 1980s. In the earliest example, antisense oligonucleotides (ODNs) covalently attached to a poly(l-lysine) backbone were investigated for their ability to inhibit the synthesis of vesicular stomatitis virus proteins; and indeed it was found that these conjugates acted as antiviral agents (1). With this as a starting point, various applications

Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_15, © Springer Science+Business Media, LLC 2011

239

240

Zimmermann et al.

employing these types of bioorganic hybrid structures have been realized. Within the field of diagnostics, molecular beacons have been constructed using negatively charged, highly emissive conjugated polymer units instead of small molecule emitters. These macromolecular probes have been demonstrated to be particularly well suited for nucleic acid detection (2). In the context of biomolecular separation, using a different combination of nucleic acids with synthetic polymers can lead to important effects on the physical properties of conjugate materials. For example, poly(N-isopropylacrylamide) (PNIPAM) is a polymer that exhibits a remarkable phase transition in aqueous medium in response to temperature changes, which is characterized by a lower critical solution temperature. This fully reversible temperature-responsive behavior has been exploited in applications where PNIPAM was covalently connected to DNA. In one report, plasmid DNA was purified by triple helix formation with a DNA-PNIPAM conjugate and subsequent precipitation at 40°C (3). Moreover, similar types of thermoresponsive conjugates have also been employed for the selective separation of DNA binding proteins (4). Another very important class of nucleic acid-polymer hybrid structures in the field of biomedicine is conjugates of ODNs and poly(ethylene glycol) (PEG). While aptamers, antisense ODNs, and small interfering RNA (siRNA) are all attractive drug candidates because of their extremely high selectivity in target recognition, they suffer from several drawbacks, including low stability against nucleases, fast excretion, and poor cellular uptake. One way to tackle these shortcomings is the “PEGylation” of pharmaceutically active nucleic acids. One excellent example of such an approach is a PEGylated 28-mer aptamer known as pegaptanib, which was approved by FDA for the treatment of age-related macular degeneration of the retina (5) and commercialized in 2004 under the brand name Macugen. In the era of nanomedicine, PEGylated poly(nucleic acids) were also transformed into nano-objects for the purpose of improving the delivery of antisense ODNs (6, 7) and siRNA (8) to cellular targets. The generation of nanosized particles was achieved by complexation with positively charged polyelectrolytes, and the resulting polyelectrolyte micelles exhibited a core/shell structure. Within these nanostructures, ionic interactions between the nucleic acid moiety and the positively charged polymer mediate the formation of the core region while the surrounding shell is composed of the neutral PEG moiety. Other types of hybrid nanostructured particles have also been fabricated by the self-assembly of amphiphilic DNA block copolymers (DBCs). In contrast to polyelectrolyte micelles, the micelles produced by this approach are comprised of a hydrophobic polymer core and a nucleic acid shell. Due to spontaneous microphase separation, the nanoparticle formation process is both facile and

Amphiphilic DNA Block Copolymers

241

convenient, and offers numerous possibilities for structural modification within the particle core as well as in the singlestranded (ss) DNA corona, which results in a plethora of different potential applications. Indeed, this important type of DNApolymer nanomaterial will be the subject of this chapter. The synthesis and characterization of these bioorganic hybrid structures will be described first (Subheading 3.1–3.3). Subsequently, methods for the investigation of the structural properties of nanoscopic DBC aggregates will be discussed (Subheading 3.4). Finally, in the last part of this chapter, general procedures will be presented for the functionalization of DBC nano-objects (Subheading 3.5). In contrast to nucleic acid-polymer hybrids containing water-soluble organic polymers (9), amphiphilic DBCs require a special preparation technique since both constituent components have different solubility properties and, therefore, cannot be coupled in aqueous phase. DBC structures with a hydrophobic polymer unit were synthesized by a grafting-onto approach on the solid phase (10). As long as the nucleotides are protected on the solid support, they react well with hydrophobic polymers in organic solvents. The key reaction intermediates were phosphoramidite polymers, which could be coupled to the 5¢ ends of ODN chains attached to the solid-phase support. An outstanding characteristic of this synthetic procedure is that the DBCs were prepared in a single, fully automated process using a DNA synthesizer in quantities of up to several hundred milligrams. Furthermore, DBCs can be generated with either a relatively short DNA block (up to 60 nucleotides (nt)) or an extended nucleic acid segment (>1,000  nt). To demonstrate the latter, the polymerase chain reaction (PCR) – a common technique employed in molecular biology as well as in paternity and diagnostic testing applications – was recently used for the first time as a synthetic tool to produce DNA-polymer hybrids; however, DBCs with high molecular weight (>25 kDa) nucleic acid units will not be described here – for additional information, the reader is referred to the relevant literature (11, 12). After establishing effective synthetic routes for DBCs, the structural properties of these materials need to be studied. Amphiphilic DBCs self-assemble into spherical micelles in aqueous solution, as visualized by atomic force microscopy (AFM) and measured by dynamic light scattering (DLS) and fluorescence correlation spectroscopy (FCS) (10). One of the aims in working with such materials is to manipulate the morphologies of the DNA block copolymer aggregates through mild stimuli such as molecular recognition processes induced by, for example, the addition of complementary ODNs. In a recent study, the hybridization of single-stranded DNA block copolymer micelles with long DNA templates containing multiple copies of the micelle

242

Zimmermann et al.

corona complementary sequence was shown to induce the transformation of the spherical micelles into rod-like aggregates (13). In this case, the Watson–Crick motif of base pairing aligned the hydrophobic polymer segments along the DNA double helix, which then resulted in selective dimer formation between the hybridized DNA block copolymer structures. This study for the first time demonstrated that DNA nanostructures, which are typically generated through the base pairing of complementary ODN sequences, can be built up by employing hydrophobic interactions, adding a new tool to the field of nanotechnology for creating novel nanostructures. DBC-based materials can serve not only as tunable, responsive nanoscale objects, but also as three-dimensional nanoscopic scaffolds for organic reactions. In one study conducted by our group (10), the DNA strands of the corona were organized by the hydrophobic interactions of the organic polymer segments in such a fashion that several DNA-templated organic reactions could be performed in a sequence-specific manner both at the surface of the micelles and at the interface between the biological and the organic polymer blocks. These results demonstrated that DNA block copolymer micelles can thus be regarded as programmable nanoreactors. Moreover, DBC self-assembled scaffolds have also been applied for enzymatic reactions (14). By incubating the DBC aggregates with a template-independent DNA polymerase, facile control over the growth of the nanoparticles under mild isothermal conditions in an aqueous medium was achieved. It was possible to gradually increase the micelle size by a factor of up to 2.4, thereby demonstrating control over nanoparticle size by an enzymatic reaction for the first time. In a further step, DBC aggregates have been investigated for their interactions with living cells. Since spherical and rod-shaped DNA nanoparticles composed of exactly the same materials were readily available, we assessed whether different geometries can influence cellular uptake. Interestingly, it was found that the rodshaped nanoparticles were taken up by cells 12 times more efficiently than their spherical counterparts (15). After looking at nondirected internalization, the targeted uptake of DNA nanoobjects was also studied. For this purpose, DBC aggregates were equipped with folate as a targeting ligand for cancerous cells. The functionalization of the particles was carried out in a very straightforward manner, i.e. ligand–ODN conjugates were hybridized with the micelles. The incorporation of fluorescent reporter groups by the same procedure revealed that receptor-mediated endocytotic uptake of the nanoparticles with a diameter of approximately 10 nm was most efficient when the maximum number of ligands was present on the rim of the micelles. The loading of doxorubicin (DOX), an anticancer drug, into the hydrophobic interior of the ligand-containing micelles resulted in efficient cytotoxicity and

Amphiphilic DNA Block Copolymers

243

high mortality among the targeted cancerous cells. The advantage of using DBC carriers compared to other types of polymeric carriers such as dendrimers or conventional block copolymer micelles is that functionalization and size can be altered extremely easily, allowing for the combinatorial testing of various DBC-based drug delivery systems (16). In the following sections, we describe procedures for the synthesis and characterization of multifunctional DBC nanoparticles such as those described above.

2. Materials 2.1. Synthesis of DNA-b-PPO

1. Poly(propylene oxide) monobutyl ether (6,800 Da) (SigmaAldrich, Germany).

2.1.1. Synthesis of the PPO Phosphoramidite (4)

2. 2-Cyanoethyl N,N-diisopropylchlorophosphoramidite (1  g ampoule) (Sigma-Aldrich). Store at −20°C. 3. Dichloromethane (DCM). “Extra dry” quality, with water content less than 0.005% (ACROS organics). 4. N-Ethyldiisopropylamine (DIPEA). Water content less than 0.3%, stored over molecular sieves (Sigma-Aldrich). 5. Molecular sieves (3 Å). 6. Sodium carbonate solution: 5% (w/v) sodium carbonate in water. 7. Brine: Saturated solution of sodium chloride in water. 8. Magnesium sulfate. 9. Rotary evaporator. 10. Glassware and other materials: Three-neck flask (100  mL), septum, bubble counter, magnetic stirrer, stir bar, 1  mL syringe, separation funnel (250  mL), round-bottom flask (250 mL), Parafilm, and NMR tube. All glassware should be dry and clean. 11. Nuclear magnetic resonance (NMR) spectra (31P) were recorded on a Bruker DRX 500 NMR spectrometer. 12. Deuterated solvent for NMR measurements: Tetrahydrofuran (THF)-d8 (Sigma-Aldrich).

2.1.2. DNA Block Copolymer Solid-Phase Synthesis

1. DNA synthesizer: ÄKTA oligopilot 100 plus DNA synthesizer (GE Healthcare) loaded with Unicorn 5.11 control software (GE Healthcare). Synthesis scale: 260 mmol. 2. Solid support: Polystyrene-based dA Primer Support (GE Healthcare). Loading: 200 mmol/g. 3. Nucleoside phosphoramidites: 0.1 M solution in anhydrous acetonitrile (ACN) (Proligo).

244

Zimmermann et al.

4. Poly(propylene oxide) monobutyl phosphoramidite: 0.1  M solution in anhydrous DCM. 5. Activator: 0.3 M solution of 5-(benzylmercapto)-1H-tetrazole (emp Biotech GmbH, Germany) in anhydrous ACN. 6. Oxidizing reagent: Solution of iodine (0.05  M) in water/ pyridine, 90:10 (v/v) (Merck). 7. Detritylation (Proligo).

reagent:

Dichloroacetic

acid

in

toluene

8. Capping A: N-Methylimidazole (20%, v/v) in anhydrous ACN (Capping A, Proligo). 9. Capping B: Acetic anhydride in anhydrous ACN (Capping B1, Novabiochem) and 2,6-lutidine in anhydrous ACN (Capping B2, Novabiochem). Both solutions are mixed (1:1, v/v) to give the final Capping B solution (acetic anhydride (20%) and 2,6-lutidine (30%) in ACN), which is connected to the DNA synthesizer. 10. Diethylamine (DEA) solution: 20% (v/v) in anhydrous ACN (Proligo). 11. ACN: Water content <30 ppm (Proligo). 12. Molecular sieves (3 Å). 13. Ethanol/water solution (1:1, v/v). 14. Deprotection solution: Concentrated ammonia. 15. Vacuum pump with pressure controller, membrane filter (0.2  mm) (Whatman), and filter holder for screw thread (GL 45). 2.2. Purification of DBCs 2.2.1. Preparative Denaturing Polyacrylamide Gel Electrophoresis

1. Acrylamide solution: 40% (w/v) acrylamide/bisacrylamide (19:1, w/w), electrophoresis grade (FisherBioReagents). Caution: Unpolymerized acrylamide is toxic and is a probable human carcinogen. Take care when handling – wear nitrile gloves and exchange gloves immediately when contaminated with acrylamide solution. 2. Water: Milli-Q (Millipore Inc., USA) or other water of comparable quality; typical resistivity is 18.2 MW/cm. 3. Urea: 99.5%, “for molecular biology” grade; DNAse-, RNAse-, and protease-free. 4. Ammonium persulfate (APS) solution: 3% (w/v) solution in water. 5. N,N,N ¢,N ¢-Tetramethylethylenediamine (TEMED). Store at 4°C. 6. 10× TBE buffer: In a volumetric flask, dissolve 108  g of tris(hydroxymethyl)aminomethane (Tris) (“for molecular biology” grade; DNAse-, RNAse-, and protease-free), 55 g

Amphiphilic DNA Block Copolymers

245

of boric acid (“for molecular biology” grade), and 7.4 g of ethylenediaminetetraacetic acid (EDTA) in ~700  mL of water. Add water to a final volume of 1 L. 7. Loading buffer: A solution of urea in 1× TBE buffer (saturated in the cold) can be used. As a visual reference for the progress of electrophoresis, xylene cyanol and bromophenol blue (e.g., 0.015%, w/v) can be added to an aliquot of loading buffer. A few microliters of this dye solution can be loaded into a separate well of the gel. 8. Thin layer chromatography (TLC) plate: Silica gel, 20 × 20 cm, with fluorescence indicator (Merck, Germany), covered with plastic wrap. Store in the dark. 9. UV lamp (l = 254 nm). 10. NAP™ desalting/buffer exchange columns (e.g., NAP™-25) (GE Healthcare). 11. Dialysis tubing: Spectra/Por 7 (3.5-kDa MWCO) (Spectrum Laboratories, USA). 12. UV–visible spectrophotometer. 2.2.2. Preparative Anion Exchange Chromatography

1. ÄKTA explorer or ÄKTA Purifier FPLC system (GE Healthcare). 2. Anion exchange column: HiTrap Q HP (5  mL) (GE Healthcare). 3. Desalting columns: Healthcare).

HiTrap

desalting

(5  mL)

(GE

4. Buffer A: 25 mM Tris–HCl, pH 8.0. 5. Buffer B: 25 mM Tris–HCl, pH 8.0, 1.0 M NaCl (analytical grade). 6. Ultrapure water (typical resistivity: 18.2 MW cm). 7. Vacuum pump with pressure controller, membrane filter (0.23  mm) (Whatman), and filter holder for screw thread (GL 45). 2.3. Characterization of DNA Block Copolymers 2.3.1. Analytical Denaturing Polyacrylamide Gel Electrophoresis

1. 10× TBE buffer, acrylamide solution, urea, APS solution and TEMED are used as described in Subheading 2.2.1 above. 2. Loading buffer: Combine one volume of cold saturated urea solution in 1× TBE buffer with an equal volume of glycerol/ water (1:1, v/v). Add bromophenol blue and xylene cyanol (e.g., 0.015%, w/v). 3. Staining solution: Dilute 10 mL of SYBR® Gold Nucleic Acid Gel Stain stock solution (Invitrogen) in 100 mL of 1× TBE buffer (see Note 14). Store in the dark.

246

Zimmermann et al.

2.3.2. Matrix-Assisted Laser Desorption/ Ionization-Time of Flight Mass Spectrometry

1. Matrix-assisted laser desorption/ionization-time of flight (MALDI-TOF) mass spectra were recorded on a Bruker MALDI-TOF mass spectrometer (Reflex TOF). 2. Matrix solution: Prepare a solution containing 20  mg of 3-hydroxypicolinic acid, 2 mg of picolinic acid, and 3 mg of ammonium citrate in 0.5 mL water/ACN (7:3, v/v). 3. Cation exchange resin: Dowex® 50  W X8 (100 mesh, ammonium-form).

2.3.3. Analytical Anion Exchange Chromatography

1. ÄKTA explorer or ÄKTA purifier FPLC system (GE Healthcare). 2. Buffers A and B are prepared as described in Subheading 2.2.2 above. 3. Anion exchange column: HiTrap Q HP (1  mL) anion exchange column (GE Healthcare). 4. NAP™-5 desalting/buffer Healthcare).

2.4. Investigation of Self-Assembled Aggregates from DNA Block Copolymers

exchange

column

(GE

1. Zetasizer 5000 (Malvern Instruments Ltd, UK). 2. Milli-Q water (Millipore Inc.). 3. Plastic cuvettes: 10-mm light path (Dispolab, Switzerland).

2.4.1. Dynamic Light Scattering 2.4.2. Fluorescence Correlation Spectroscopy

1. TAE buffer: 20 mM Tris–HCl, 10 mM acetic acid, 0.5 mM EDTA, pH 8.0, and containing 100 mM NaCl and 20 mM MgCl2. 2. Alexa Fluor 488- or other fluorophore-labeled complementary DNA. 3. Glass microscope slides: 25 × 75 × 1.2 mm.

2.4.3. Atomic Force Microscopy

1. Multimode scanning probe microscope equipped with a Nanoscope IIIa controller (Veeco Instruments). 2. Immobilization buffer: 50 mM magnesium acetate (MgAc2) diluted with Milli-Q water from a 1 M MgAc2 stock solution (Sigma-Aldrich), sonicated and passed through a 0.23  mm syringe filter device. 3. Sample substrates: 9.5-mm V-1 Muscovite mica disks and 15-mm metal specimen disks (both from Electron Microscopy Sciences). 4. Cantilevers: ACTA silicon cantilevers ( f0 = 200–400  kHz, k = 25–75 N/m) (AppNano, USA) for imaging in air; NPS or SNL silicon nitride cantilevers ( f0 = 40–75 kHz, k = 0.32 N/m) (Veeco Probes, USA) for imaging in fluid.

Amphiphilic DNA Block Copolymers

2.5. Drug Loading of DNA Block Copolymer Nanoparticles and Surface Functionalization by Hybridization

247

1. Doxorubicin (Sigma-Aldrich). 2. Tetrahydrofuran. 3. Centrifugal filter devices (100 kDa MWCO) (Millipore). 4. Syringe filter devices (0.45 mm) (VWR). 5. Sonicator.

2.5.1. Loading of DBC Nanoparticles with Hydrophobic Compounds 2.5.2. Surface Functionalization of DBC Nanoparticles by Nucleic Acid Hybridization

1. Amino-modified ODN: 5¢-NH2-TAACAGGATTAGCAGA GCGAGG-3¢, MW = 6,950 Da. 2. Folic acid. 3. 4-(4,6-dimethoxy-1,3,5-triazin-2-yl)-4-methylmorpholinium chloride (Sigma-Aldrich). 4. TAE buffer: 20 mM Tris–HCl pH 8.0, 10 mM acetic acid, and 0.5 mM EDTA containing 100 mM NaCl and 60 mM MgCl2.

3. Methods 3.1. Synthesis of DNA-b-PPO

The following protocol describes, as an example, the synthesis of a DNA-b-PPO molecule consisting of a 6.8 kDa poly(propylene oxide) monobutyl ether block connected to the 5¢-end of a 22-mer ODN (5¢-CCTCGCTCTGCTAATCCTGTTA-3¢) through a phosphate diester linkage. This synthesis protocol, as presented herein, was originally designed for use with an ÄKTA oligopilot 100 DNA synthesizer on a relatively large synthesis scale (120  mmol); it should, however, be possible to downscale and modify the protocol for the synthesis of DNA-b-PPO with a small-scale DNA synthesizer. The DNA poly(propylene oxide) block copolymer is accessible via a grafting-onto procedure in-line on a DNA synthesizer using a solid-support technique and phosphoramidite chemistry (Fig. 1). Initially, the ODN is automatically synthesized on a solid support following standard protocols. Subsequently, the poly(propylene oxide) phosphoramidite 3, which must be prepared in advance, is coupled to the free 5¢-hydroxy function of the ODN. Cleavage from the support and deprotection of the nucleobases of the synthesized DNA block copolymer 4 is then achieved again according to standard protocols.

3.1.1. Synthesis of the PPO Phosphoramidite (4)

The phosphoramidite of the poly(propylene oxide) is accessible through the reaction of the hydroxy-terminated poly(propylene oxide) monobutyl ether 1 with the phosphitylation reagent,

248

Zimmermann et al.

Fig. 1. Conversion of the hydroxy-terminated poly(propylene oxide) monobutyl ether 1 and phosphitylating reagent 2 into the corresponding phosphoramidite 3, which is then coupled to the 5¢-hydroxy end of the oligonucleotide on a DNA synthesizer. Deprotection of the phosphates and nucleobases, followed by cleavage from the solid support, yields the product DNA-b-PPO 4.

2-cyanoethyl N,N-diisopropylchlorophosphoramidite 2, in the presence of DIPEA according to the following procedure: 1. Prepare a dry three-neck flask equipped with a magnetic stirring bar, argon supply, bubble counter, and a septum. Dissolve 13.6 g (2 mmol) of poly(propylene oxide) monobutyl ether (1) and 442 mL (0.33 g, 2.6 mmol, 1.3 eq.) of dry DIPEA in approximately 50 mL of dry DCM (see Note 1). 2. Open a fresh ampoule of the phosphitylating agent, 2-cyanoethyl N,N-diisopropylchlorophosphoramidite (2), transfer 580  mL (0.615  g, 2.6  mmol, 1.3  eq.) into a syringe, and add the phosphitylating reagent dropwise to the stirred solution (via the septum). Let the reaction solution stir for another 3 h at room temperature. 3. Transfer the reaction mixture into a separation funnel and wash consecutively with sodium carbonate solution (5%, w/v), brine (three times), and then finally with water (three times). 4. Transfer the organic phase into an Erlenmeyer flask equipped with a stirring bar and add magnesium sulfate (until “fluffy flakes” of the salt are observable) to remove residual water. 5. Filtrate the suspension (using a folded filter) to remove the magnesium sulfate. Rinse the magnesium sulfate with a few milliliters of dry DCM. Collect the filtrate in a dry roundbottom flask (see Note 2).

Amphiphilic DNA Block Copolymers

249

6. Evaporate to dryness on a rotary evaporator and, if possible, flush the rotary evaporator with argon instead of air (see Note 3). 7. Dry the residue under high vacuum for at least 16 h. Flush the round-bottom flask with argon, close it with a glass stopper, and seal with parafilm. Store the phosphoramidite at −18°C (see Note 4). Yield: approximately 90%. 8. Characterization of the phosphoramidite: Dissolve approximately 30 mg of the phosphoramidite in 750 mL of deuterated THF in an Eppendorf tube. Transfer this solution into a dry and argon-flushed NMR tube, close it and seal with parafilm. Perform a 31P-NMR measurement (broadband proton decoupled) using triphenylphosphine in THF as a standard. A signal at d = 145.6 ppm indicates the successful conversion of the poly(propylene oxide) monobutyl ether 1 into the phosphoramidite 4 (see Note 5). 3.1.2. DNA Block Copolymer Solid-Phase Synthesis

3.1.2.1. Presynthesis Preparations

The procedure described below for solid-phase DNA block copolymer synthesis is for a 120-mmol synthesis scale using an ÄKTA oligopilot 100 plus DNA synthesizer. The synthesis of the 22-mer ODN, 5¢-CCTCGCTCTGCTAATCCTGTTA-3¢, is carried out by employing standard parameters for the nucleoside phosphoramidite coupling routines with a 6.3-mL column reactor provided by the manufacturer. The synthesis method is designed, controlled, and monitored using the Unicorn 5.11 control software. 1. Presynthesis purging is performed to freshen all chemicals in the tubing lines of the DNA synthesizer (see Note 6). 2. 1.3 g of dA solid support is weighed and transferred into the column reactor. 3. The column reactor is mounted to the column holder of the DNA synthesizer.

3.1.2.2. Synthesis Procedure

A standard nucleoside phosphoramidite synthesis cycle in the order detritylation, coupling, oxidation, and capping is performed as recommended by the manufacturer. For process control, the column pressure, conductivity, and UV absorbance at 280 and 350  nm are monitored during the entire synthesis procedure. The detritylation area (unit: mAU × mL) is used as a measure of the coupling efficiency. 1. A volume of 50 mL of ACN is passed through the column reactor to pre-swell the solid support. 2. Standard DNA ODN synthesis cycle: After the detritylation step, 1.5  eq. of the nucleoside phosphoramidite are passed through the column reactor and further recycled for 3 min (coupling and recycling steps). Subsequently, the oxidation and capping steps are carried out.

250

Zimmermann et al.

3. Polymer synthesis cycle: After the detritylation step, 4.0 eq. of the PPO phosphoramidite 4 are passed through the column reactor and further recycled for 30 min (coupling and recycling steps). Subsequently, the oxidation step is carried out. 4. Finally, treatment with DEA solution (40 mL in 10 min) is performed (see Note 7). 3.1.2.3. Postsynthesis Procedure

1. The solid support in the column reactor is dried in vacuo. 2. The dry support is transferred to a sealable container (e.g., 100-mL glass bottle with screw cap) and reswelled with less than 1 column volume (<6.3 mL) of ACN. 3. Perform following steps in a fume hood: A volume of 40 mL of concentrated ammonia is added to the solid support and the bottle is sealed with plastic tape. The deprotection mixture is incubated for 16 h at 55°C. 4. The deprotection mixture is cooled to room temperature (see Note 8) and the solid support is removed by filtration (glass filter, pore size 3). Residual material is washed from the solid support with 100 mL of ethanol/water (1:1, v/v). The filtrate is evaporated carefully (see Note 9) on a rotary evaporator.

3.2. Purification of DNA Block Copolymers

The coupling reactions on the DNA synthesizer have efficiencies in the range of 98–99% but are not perfectly quantitative. As a result, to a small extent, all possible deletion mutants (viz., ODNs with N-1, N-2, N-3, … nucleotides) of the 22-mer sequence are also generated during the synthesis procedure. The full-length DNA block copolymer can be separated from these by-products either by preparative denaturing polyacrylamide gel electrophoresis (PAGE) or by anion exchange chromatography.

3.2.1. Preparative Denaturing Gel Electrophoresis

The DNA block copolymer is most effectively purified in 15% (w/v) polyacrylamide gels containing 7 M urea and using 1× TBE as the running buffer. Instructions for the preparation of the polyacrylamide solution are presented below for casting two preparative gels of the following dimensions: 170 × 163 × 2 mm. This protocol can be easily adapted to prepare gels of any other size, as well. 1. Prepare a 2-mm thick denaturing 15% (w/v) polyacrylamide gel by mixing 53 mL of acrylamide solution (40% w/v, acrylamide/bisacrylamide 19:1 w/w), with 14  mL of 10× TBE buffer and 16.6 mL of water. Add to the stirred solution 58 g of urea (see Note 10). 2. Assemble the clean glass plates and spacers according to the manufacturer’s instructions. Set aside a comb with one large pocket (for loading the DNA block copolymer) and one small pocket (for loading a reference and/or loading dye) ready for use.

Amphiphilic DNA Block Copolymers

251

3. Add 168 mL of TEMED and 5.6 mL of APS solution to the acrylamide gel premix. 4. Pour the gel and insert the comb into the gel cassette. Avoid generating air bubbles while casting the gel, especially at areas near the comb’s teeth. The polymerization of the polyacrylamide takes approximately 60 min. 5. Prepare the running buffer, 1× TBE, by diluting 200 mL of 10× TBE buffer with 1,800  mL of water in a measuring cylinder and mix thoroughly. 6. Wash the gel cassette with water to remove urea and gel residues from the glass plates. Remove the comb carefully and flush the wells with distilled water. 7. If available, start the cooling module of the electrophoresis apparatus. Insert the gel cassette into the electrophoresis cell and add running buffer to the upper and lower buffer reservoirs. Ensure that there is no buffer leaking from the upper buffer reservoir. Complete the assembly of the electrophoresis cell and connect the electrodes to the power supply. Turn the electrophoresis apparatus on and let it run at a voltage of 450  V for approximately 15  min to ensure that an equal temperature has been reached throughout the gel. 8. Dissolve a maximum of 100 nmol of the DNA block copolymer in 200  mL of loading buffer (see Note 11). Xylene cyanol and bromophenol blue can be used to monitor the progress of the electrophoresis run; depending on the acrylamide concentration used, the electrophoretic mobilities of xylene cyanol and bromophenol blue are comparable to 12-mer/8-mer (10/20% PAGE gel, 7 M urea) ODNs and 55-mer/28-mer (10/20% PAGE gel, 7  M urea) ODNs, respectively (17). 9. Flush the wells thoroughly with an excess of running buffer by using a syringe. 10. Load the gel with the samples and run at 450 V for 4–5 h. 11. Stop the electrophoresis and remove the gels from the electrophoresis cell. Remove the spacers and carefully lift off one of the glass plates. Transfer the gel onto a TLC plate (with fluorescence indicator) covered with plastic wrap. Under UV light, the DNA bands will become visible against the TLC plate. Note that since UV light is absorbed by the ODNs, the nucleic acid bands will, therefore, appear dark in the gel (this process is called UV-shadowing). 12. Cut out the band with the lowest electrophoretic mobility using a razor blade or a scalpel and transfer the gel pieces into a 15-mL Falcon conical tube. 13. Crush the gel mechanically (e.g., by using a Teflon rod) and add two to three volumes of water. Place the Falcon tube on a

252

Zimmermann et al.

shaker and extract the gel overnight at room temperature. Centrifuge for 20 min at 4,000 × g and decant the supernatant. Extract one more time with water and combine the extracts. 14. Evaporate the extract to dryness by using a vacuum centrifuge, and redissolve the residue in water (the required volume depends on the amount of residual buffer and urea present). 15. Desalt the crude product using either a NAP™ desalting column or dialysis tubing (3.5 kDa MWCO). 16. Lyophilize the desalted product and redissolve it in Milli-Q water. 17. The concentration of the resulting DNA block copolymer solution can be estimated by measuring the absorbance at 260  nm using a UV–visible spectrophotometer (1-cm light path quartz cuvette) (see Note 12). 3.2.2. Preparative Anion Exchange Chromatography

The following instructions for preparative anion exchange chromatography of the DNA block copolymer assume the use of an ÄKTA explorer FPLC equipped with a 5-mL HiTrap Q HP column. During elution, the absorbance of the ODN at 260 nm is monitored. A linear gradient of Buffer B is applied for resolving ODNs having a different number of charges. Though this signal may be saturated at the concentrations described here, the product (retention time » 40 min) may nevertheless be separated from pristine (unconjugated) DNA (retention time » 32 min) and other side products. 1. Buffers A and B are degassed through a membrane filter under low pressure (60 mbar) prior to use. 2. The dry crude obtained from Subheading 3.2.1 is dissolved in Buffer A (50 mL) and filtered through a 0.23-mm syringe filter. The filtered crude mixture is then connected to the FPLC sample inlet. 3. The crude mixture (2 mL) is injected onto the HiTrap Q HP column via the sample valve. 4. After the injection, a linear salt gradient is employed to isolate the product: Start from 0% NaCl (0% Buffer B) and ramp up at a slope of 16 mM NaCl per minute (flow rate: 5 mL/min) to 0.8 M NaCl (80% Buffer B). 5. The peak with maximal retention time, which elutes at approximately 40 min, is collected. 6. Steps 4 and 5 are repeated (in an automated manner using the batch processing software) until all of the crude mixture is consumed. 7. The collected fractions containing DNA-b-PPO are concentrated on a rotary evaporator to reduce the volume for the subsequent desalting step (see Note 9).

Amphiphilic DNA Block Copolymers

253

8. The DBC solution is desalted through HiTrap desalting columns (2 × 5 mL columns are connected in series, thus the total column volume is 10 mL) using Buffer A. 9. Step 8 is repeated again in an automated manner (see Note 13). 10. The DNA-b-PPO fraction is carefully concentrated on a rotary evaporator by gradually decreasing the pressure (see Note 9). 11. The concentrated solution is further dialyzed to remove components from the elution buffer. The material is dialyzed (3.5 kDa MWCO) seven times against 2 L of water. 12. After dialysis, the DBC solution is lyophilized and redissolved in Milli-Q water. 13. The concentration of the purified DNA-b-PPO solution is determined by measuring the absorbance at 260 nm using a UV–visible spectrophotometer (1-cm light path quartz cuvette) (see Note 12). 3.3. Characterization of DBCs

The characterization of the synthesized DNA block copolymer is best achieved by using a combination of electrophoretic, chromatographic, and mass spectrometric analyses. Gel electrophoretic analysis of the DNA block copolymer reveals a reduced electrophoretic mobility compared to unmodified ODNs of identical sequence and/or ODNs of similar length from a DNA ladder (Fig. 2). The reduced electrophoretic mobility of the DNA block copolymer can be explained by the increased mass-to-charge ratio, the hydrophobic nature of the conjugated poly(propylene oxide) and presumably the morphology of the hybrid molecule. The DNA block copolymer can also be further characterized by MALDI-TOF mass spectrometry. This relatively soft ionization technique allows analyses of the large DNA block copolymer to be performed without fragmentation (for example, see Fig.  3). And finally, analysis of the DNA block copolymer by means of

Fig. 2. Denaturing PAGE analysis of crude (a) and gel purified DNA-b-PPO (b). Lane 1: Purified DNA-b-PPO. Lane 2: Unreacted DNA oligonucleotide.

254

Zimmermann et al.

Fig. 3. MALDI-TOF mass spectrum of DNA-b-PPO (m/zfound: 13,593 Da; m/zcalcd. = 13,600 Da). The spectrum was recorded in negative ion mode using 3-hydroxypicolinic acid as the matrix.

Fig.  4. Chromatogram from analytical anion exchange chromatography. (a) Purified DNA-b-PPO. (b) Unmodified DNA oligonucleotide.

Amphiphilic DNA Block Copolymers

255

anion exchange chromatography shows an increased retention time compared to a reference ODN of the same sequence due to the additional phosphodiester bond connecting the DNA and poly(propylene oxide) block (Fig. 4). 3.3.1. Analytical Denaturing Gel Electrophoresis

1. Prepare a denaturing 20% (w/v) polyacrylamide gel. To cast a 100 × 90 × 1 mm gel, combine 6.4 mL of acrylamide solution (40% w/v, acrylamide/bisacrylamide 19:1 w/w) with 1.25 mL of 10× TBE buffer in an Erlenmeyer flask equipped with a magnetic stir bar. Add 5.5 g of urea to the stirred solution (see Note 10). 2. Assemble clean glass plates and spacers according to manufacturer’s protocols. 3. Add 15 mL of TEMED and 0.5 mL of APS solution. 4. Pour the gel and insert the comb into the gel cassette. Avoid generating air bubbles in the gel, especially at areas near the comb’s teeth. The polymerization of the polyacrylamide takes approximately 60 min. 5. Prepare the running buffer, 1× TBE buffer, by diluting 100 mL of 10× TBE with 900 mL of water in a measuring cylinder, and mix thoroughly. 6. Wash urea and gel residues from the glass plates. Insert the comb carefully and flush the wells with distilled water. Insert the gel cassette into the electrophoresis cell and add running buffer in the upper and lower buffer reservoirs. Ensure that no buffer is leaking from the upper reservoir. 7. Connect the electrodes to the power supply. Turn the electrophoresis apparatus on and let it run at a voltage of 100 V for approximately 15 min to ensure that an equal temperature has been reached throughout the gel. 8. Dissolve 10–30 pmol of the DNA block copolymer in loading buffer. As a reference, also prepare solutions with equal amounts of a suitable DNA ladder and an ODN of the same sequence as in the DNA block copolymer. Xylene cyanol and bromophenol blue can be used to monitor the progress of electrophoresis: The electrophoretic mobilities of xylene cyanol and bromophenol blue in a 20% (w/v) denaturing PAGE are comparable to 8-mer ODNs and 28-mer ODNs, respectively (17). Prior to loading, heat the samples to 90°C for 5  min to disrupt any nonspecific binding/hybridization between subsequences on the DBCs. 9. Flush the wells thoroughly with an excess of running buffer by using a syringe. 10. Load the gel with the samples and run for 1 h at 100 V.

256

Zimmermann et al.

11. Prepare the staining solution in a suitable container (e.g., polypropylene container) by diluting 10  mL of SYBR® Gold Nucleic Acid Gel Stain in 100 mL of 1× TBE buffer (see Note 14). 12. Once the electrophoresis run has completed, open the gel cassette and carefully remove one of the glass plates while keeping the gel in place on the other plate. Place the gel together with the glass plate into the staining solution with the gel side up. A sufficient level of staining is usually achieved after 15–20 min. 13. Remove the glass plate with the gel from the staining solution, rinse with a few milliliter of water, and transfer the gel to a gel imaging system. An example gel image is shown in Fig. 2. 3.3.2. MALDI-TOF Mass Spectrometry

1. For MALDI-TOF analysis, the DNA-b-PPO is desalted to obtain sharp signals. For this purpose, a desired amount of the DNA-b-PPO material is subjected to repeated NAP™ column purification. 2. The desalted sample solution is evaporated, and water is added to the residue to obtain a DNA-b-PPO concentration of 50–100 mM. 3. A few microliters of the DNA-b-PPO solution are combined with an equal volume of matrix solution, and then spotted onto the sample plate according to the manufacturer’s instructions (see Note 15). 4. MALDI-TOF MS measurement is performed in negative ion mode.

3.3.3. Analytical Anion Exchange Chromatography

An ÄKTA Explorer or Purifier FPLC system equipped with a 1-mL HiTrap Q HP anion exchange column is used for analytical chromatography of the synthesized DNA-b-PPO. During the elution of the DBC sample, the absorbance at 260  nm is ­monitored. Buffers A and B are prepared as described in Subheading 2.2.2. 1. A linear salt gradient with a slope of 16 mM NaCl per minute is employed during elution (flow rate: 1 mL/min). Start from 0% NaCl (0% Buffer B) and ramp up to 0.8  M NaCl (80% Buffer B). 2. The two peaks with the longest retention times, assigned as DBC and unmodified 22-mer ODN, are collected and can be analyzed by gel electrophoresis for further confirmation. 3. Prior to gel analysis, the samples are desalted using a NAP™-5 nucleic acid purification column.

Amphiphilic DNA Block Copolymers

257

An example chromatogram from analytical anion exchange chromatography of purified DNA-b-PPO is shown in Fig. 4. 3.4. Investigation of the Structural Properties of SelfAssembled Aggregates of DBCs

3.4.1. Dynamic Light Scattering

In our laboratory, DLS, FCS, and AFM represent the most common methods for determining the morphology of aggregates of amphiphilic DBCs. In DLS, the time dependence of intensity fluctuations in the light scattered from the sample yields information about the Brownian motion of the particles, which in turn can be related to the hydrodynamic radius (see Note 16). This technique is the least invasive of the three, and allows observation of the aggregates in solution with no modifications. FCS links the intensity fluctuations of a fluorescent dye in an extremely small volume to diffusion processes. The diffusion time varies with particle diameter, such that labeling DBCs with dye molecules allows the determination of the size of their aggregates. The third technique, AFM, requires the DBC materials to be immobilized on an atomically smooth surface. These samples are then mechanically probed with an extremely sharp tip, and surface topography is recorded. This process unavoidably leads to distortions of the particles and is by far the most invasive, but it is also the only method that permits the direct visualization of individual aggregates. 1. This protocol assumes the use of a Zetasizer 5000 DLS system. Other systems may require a different sample volume or concentration, and the following procedure can be adjusted accordingly. 2. Prepare 1  mL of a 2  mg/mL solution of DBC in Milli-Q water (see Note 17). Heat the solution to 90°C and let it cool slowly to room temperature to generate a reasonably narrowly dispersed population of aggregates at thermodynamic equilibrium. 3. Transfer the DBC solution to a disposable plastic cuvette, insert the sample into the DLS instrument, and record data at a scattering angle of 90° as instructed in the user’s manual. 4. Analyze the resulting autocorrelation function output using the DLS instrument’s analysis software or by using the CONTIN algorithm to generate the particle size distribution of the sample.

3.4.2. Fluorescence Correlation Spectroscopy

1. This protocol assumes the use of a confocal FCS system, which may either be a commercial type of setup or an inhouse built design. In particular, the FCS system should be equipped with a water immersion objective with high numerical aperture (NA > 0.9). 2. Dissolve the DBC sample in TAE buffer containing NaCl (100 mM) and MgCl2 (60 mM). The concentration of DNA in the sample should be 2–5 mM.

258

Zimmermann et al.

3. If interested in analyzing the size of the DBC micelles with the DNA corona in a single-stranded form, add dye-labeled complementary DNA at a 1% molar concentration ratio (i.e., 20–50  nM) (see Note 18). Alternatively, to study micelles with the DNA corona in a double-stranded form, add dye-labeled complementary DNA and unlabeled complementary DNA together at a molar concentration ratio of 1:99 and at a molar excess relative to the concentration of the DBC sample. 4. Heat the samples to 95°C and slowly cool to room temperature (1°C per hour) using a PCR thermocycler. 5. Place 25–50 mL of the room-temperature solution on a glass microscope slide and mount in the FCS system; collect spectra according to the user’s manual. 6. Using FCS analysis software, calculate the fluorescence intensity autocorrelation function G(tc). This can be fitted with a single diffusion time tD for the sample according to G (t c ) =

ù 1 é 1 1 é1 - T + Te -t c /t T ùû ê ú N f ë 1 + t c / t D û 1 + (w / z )2 (t c / t D ) ë

where Nf is the average number of fluorescent molecules in the confocal detection volume, tc the correlation time, w/z the ratio of the 1/e 2 radii of the detection volume in the radial and axial directions, respectively, T the average fraction of fluorophores in the triplet state, and tT the lifetime of the triplet state of the fluorophore. The ratio w/z can be measured with Rhodamine 6G or with another reference fluorophore solution. 7. The diffusion coefficient D can be related to the diffusion time by t D = w 2 / 4D and to the frictional coefficient fsphere of a sphere with radius R0 by f sphere = kT / D = 6ph R0 , where k is Boltzmann’s constant, T is the Kelvin temperature, and h is the viscosity of the solvent, thus allowing direct determination of the micelle radius from the diffusion time (see Note 19). 3.4.3. Atomic Force Microscopy

The protocols below assume the use of a Veeco MultiMode Scanning Probe Microscope (or any other similar Veeco AFM system). Other AFM systems may require different sample sizes or operate with different volumes of fluid, and the protocols can be easily scaled accordingly. The focus here is on sample preparation; imaging procedures are best described in the AFM system user’s manuals. It is possible to image DBC aggregates in air or in fluid (see Note 20 for additional details).

Amphiphilic DNA Block Copolymers 3.4.3.1. Imaging in Air

259

1. Dilute the DBC solution with 12.5 mM MgAc2 to give a final DNA concentration of 10–30 mM. 2. Attach a 9.5-mm V-1 muscovite mica disk to a 15-mm metal specimen disk using double-sided tape. Cleave the mica with a razor blade or remove the upper layers with a piece of adhesive tape to expose a fresh surface. Immediately cover with a 40-mL drop of 12.5 mM MgAc2 buffer and allow the drop to sit on the freshly cleaved mica surface for 5 min (see Note 21). 3. Gently blow the mica surface dry with a N2 stream and immediately apply 35 mL of sample solution (see Note 22). Cover and allow the sample to sit for at least 30 min, making sure that the sample does not dry out. 4. Shake off the remaining excess solution and cover the surface with a 100-mL drop of Milli-Q water for 10 s (see Note 23). Lightly shake off the water and blow dry under a gentle N2 stream. 5. Insert the sample into the AFM and calibrate according to the user’s manual. Image the sample in tapping mode using ACTA cantilevers ( f0 = 200–400  kHz, k = 25–75  N/m) mounted in a standard cantilever holder.

3.4.3.2. Imaging in Fluid

1. Prepare the sample solution and mica surface according to the air imaging protocol described above, up to the exposure of the freshly cleaved mica to MgAc2 buffer. 2. Gently blow the mica surface dry with a N2 stream and immediately apply 35 mL of sample solution (see Note 22). Allow the sample to sit for 5 min and then add 15 mL of the MgAc2 buffer used to dilute the DBC solution. 3. Insert the sample into the AFM, lower the cantilever into fluid, and calibrate according to the user’s manual. Record images in tapping mode using NPS or SNL cantilevers ( f0 = 40–75 kHz, k = 0.32 N/m) mounted in a fluid cell (see Note 24). An example of an AFM image of DNA-b-PPO micelles recorded in fluid is shown in Fig. 5.

3.5. Drug Loading of DNA Block Copolymer Nanoparticles and Surface Functionalization by Hybridization

The DNA-b-PPO micelles can be loaded utilizing both the hydrophobic character of the poly(propylene oxide) core for the incorporation of hydrophobic molecules, and the ssDNA corona for the hybridization of functionalized complementary ODNs. In the following, examples are given for the loading of the anticancer drug, DOX, into the hydrophobic PPO core, and for the decoration of the DNA corona by the hybridization of folic acid-modified ODNs (folic acid can act as a targeting ligand for cancerous cells). An in  vivo study has recently shown that the DOX-loaded and folic acid-decorated DNA-b-PPO micelles act

260

Zimmermann et al.

Fig. 5. AFM image of DNA-b-PPO micelles acquired by tapping mode operation in fluid.

as efficient drug delivery vehicles, displaying enhanced cellular uptake and high cytotoxicity (16). 3.5.1. Loading of DBC Nanoparticles with Hydrophobic Compounds

The preparation of DOX-loaded DBC micelles is performed as follows (18): 1. In a glass vial, 10 mg of the DNA block copolymer and 2 mg of DOX are dissolved in 2 mL of THF. 2. The above solution is added dropwise into a beaker containing 20 mL of water. During the dropwise addition, the beaker is sonicated. 3. The aqueous solution is exposed to air overnight (to allow evaporation of the THF solvent). 4. Residual THF is removed on a rotary evaporator. The solution is reduced to a final volume of approximately 5 mL. 5. Large polymer and DOX aggregates are removed using a syringe filter device (0.45 mm). 6. A centrifugal filter device (100  kDa MWCO) is used to remove free DOX from the solution containing drug-loaded micelles.

3.5.2. Surface Functionalization of DBC Nanoparticles by Nucleic Acid Hybridization

This section describes the hybridization of single-stranded (ss) DNA-b-PPO with a folic acid-modified complementary ODN sequence. The folic acid-modified ODN can be prepared by using different techniques; here, a method is presented that uses a standard post-synthetic modification protocol starting from a

Amphiphilic DNA Block Copolymers

261

5¢-amino-modified ODN (5¢-NH2-TAACAGGATTAGCAGA GCGAGG-3¢), which can be custom-synthesized using aminomodifier phosphoramidites available from various suppliers (see Note 25). 1. 30  mmol of the amino-modified ODN, 100  mmol of folic acid, and 35 mmol of 4-(4,6-dimethoxy-1,3,5-triazin-2-yl)-4methylmorpholinium chloride (DMT-MM) are dissolved in 1 mL of water (see Note 26). 2. The reaction solution is left to stand for 12  h at room temperature. 3. The reaction product is separated from the by-products by preparative denaturing PAGE (20%, w/v PAGE gel); the product band (i.e., the band of lowest electrophoretic mobility) is cut from the gel. The recovered gel slice is then extracted, and the extract is dialyzed against water. Finally, the solution recovered from dialysis is lyophilized to obtain the purified folic acid-modified ODN product. 4. Characterization of the purified folic acid-modified ODN is carried out by analytical denaturing PAGE (20%, w/v PAGE gel) and MALDI-TOF mass spectrometry. Example results from PAGE and MALDI-TOF MS analyses of a folic acid-modified 22-mer ODN are shown in Fig. 6. 5. The ssDNA-b-PPO is combined with the desired amount (see Note 27) of the folic acid-modified complementary ODN in hybridization buffer (20  mM TAE buffer (pH 8), containing 100 mM NaCl and 60 mM MgCl2). In our hands,

Fig.  6. (a) MALDI-TOF mass spectrum of the folic acid-modified 22-mer oligonucleotide (m/zfound: 7,385  Da, m/zcalcd.: 7,391 Da). The spectrum was recorded in negative ion mode using 3-hydroxypicolinic acid as the matrix. (b) Denaturing PAGE analysis (20% w/v polyacrylamide, 7 M urea) of the unmodified 22-mer oligonucleotide (Lane 1) and the folic acidmodified 22-mer oligonucleotide (Lane 2 ).

262

Zimmermann et al.

satisfactory results are obtained by using a final DNA-b-PPO concentration between 200 and 500 mM. 6. The hybridization solution is heated to 95°C and slowly cooled down to 4°C in a PCR thermocycler (cooling rate = −1°C per hour).

4. Notes 1. Since phosphoramidites are highly hygroscopic and tend to hydrolyze and oxidize, assure the exclusion of air and water at all times. If necessary, the argon gas can be dried by passing through a drying tower prior to use. For the phosphitylation reaction, purchase DCM and DIPEA which are specified as being “extra dry” quality. Otherwise, reflux the organic solvents over calcium hydride (approximately 3 g per 100 mL; reflux for at least 4 h) and distill under an argon atmosphere, for example, by using a dephlegmator. Store DIPEA over molecular sieves. 2. Weigh the round-bottom flask (and the glass stopper you will use to close it) before use, so that you can determine the amount of phosphoramidite in the closed flask after drying in vacuo. 3. Rotary evaporators are often shared by several lab workers; therefore, in order to avoid contamination, it is highly recommended to carefully check if the rotary evaporator is dry and clean – in particular, free from any acids and bases – prior to use. An argon balloon can be used to flush the rotary evaporator with inert gas after the evaporation of the solvents to prevent the phosphoramidite from being oxidized; however, it turns out that the crude PPO phosphoramidite (6.8 kDa PPO) is relatively stable against air oxidation, and that the argon flush is not “an absolute must” at this stage when air exposure lasts for just a few seconds. 4. Before opening the flask, always let it first warm to room temperature to avoid condensation of airborne moisture inside the flask. 5. A signal at d = 180  ppm indicates the presence of residual phosphitylating agent while signals at about d = 0 ppm indicate phosphates, oxidation products of the target compound or the phosphitylation reagent. Sometimes, the 31P NMR spectrum of the PPO phosphoramidite may show to some extent signals at 0 ppm. This can be due to oxidation of the PPO phosphoramidite sample during transfer to the NMR tube or the presence of impurities in the deuterated solvent

Amphiphilic DNA Block Copolymers

263

and may not affect the coupling efficiency. If strong signals at 0  ppm are observed, the NMR measurement should be repeated with fresh (dry) deuterated solvent using an optimized “transfer procedure.” The coupling efficiency can also be estimated by performing a test synthesis (i.e., at a small scale with a relatively short ODN). 6. Presynthesis purging is performed with the following reagents: 15 mL of solvent, capping agents, oxidizer, activator, and 2.5  mL of nucleoside phosphoramidites. Purging with the detritylation solution is performed in every synthesis method. 7. Diethylamine treatment removes the b-cyanoethyl groups from the phosphates, which can lead to N-alkylation during standard ammonia deprotection at elevated temperature. According to the manufacturer, by utilizing a diethylamine treatment, synthesis yields can be improved by 3–7%. 8. Caution: Do not open the container with the deprotection mixture before it has cooled to room temperature – otherwise, the release of ammonium gas may occur! 9. Caution: Foam formation occurs due to the amphiphilic nature of the DBCs. 10. The solvation process of the urea can be accelerated by gently (!) warming the gel solution using a water bath. When doing so, take care that the gel solution has cooled to room temperature when starting the polymerization process by adding TEMED and APS solution; otherwise, the polymerization reaction may proceed too fast to allow casting of the gel. 11. The sample loading capacity will strongly depend on the gel dimensions. Furthermore, band separation/resolution will profit from cooling during electrophoresis. This prevents the DNA bands from showing the “smiley effect,” sometimes caused by overheating of the gel. 12. The extinction coefficient of the 22-mer ODN described here is 232,400 L/mol/cm (19). If the measured optical density at 260 nm (A260) is >1, dilute the stock solution until the A260 is <1. 13. During the automated desalting procedure, the column has to be washed in water/ethanol (80:20, v/v) and equilibrated in Buffer A every ten desalting runs. The DBC fractions are combined in a large container attached to the outlet-valve. 14. Alternatively, SYBR Green II (Invitrogen), GelStar nucleic acid stain (LONZA) or ethidium bromide can be used to stain the gel. However, the efficiency of staining ssDNA with ethidium bromide is significantly lower than with the other staining reagents mentioned above.

264

Zimmermann et al.

15. Some increase in performance of the MALDI-TOF mass spectrometry can be achieved by treating the sample with ion exchange resins (ammonium-form) prior to the addition of matrix solution. Residual alkali metal cations will be exchanged by the ammonium ions. 16. The hydrodynamic radius value that is derived from DLS studies includes any solvent or other molecules that move with the aggregate; because of this, DLS typically yields larger particle size estimates than with other approaches. It should also be pointed out that this method demands far more sample material than FCS or AFM, although the material can still be used for other purposes after the DLS measurement. 17. If this quantity of material is not available, it may be possible to use a more dilute solution; however, the time required for data acquisition will increase significantly. With some materials, the signal below this concentration will be too low to yield useful data. In either case, care must be taken not to dilute the materials below the critical micelle concentration. 18. The choice of dye will depend on the excitation and emission wavelengths accessible to the specific FCS setup, but Alexa Fluor® 488- and Cy3-labeled DNA are commonly used as probes and are commercially available. The use of a 1% concentration ratio assures a sufficient dye concentration for FCS detection while maintaining the primarily ssDNA character of the micellar corona. 19. This analysis is based on the Stokes–Einstein equation and is only strictly applicable to spherical particles. 20. The micelles that remain after sample preparation for air imaging tend to collapse to some extent. Although these micelles are harder and may be imaged more aggressively than the same materials in fluid, they can also appear smaller and may aggregate during the drying process. When hydrated, these materials are particularly soft and should be imaged in soft tapping mode, in which the amplitude set point is greater than 90% of the RMS voltage. At lower amplitude set points (i.e., higher tapping forces), it is often possible to obtain stable images, but the micelles appear drastically smaller. Fluid cell imaging will give the closest possible approximation of the true size and shape of the DBC aggregates, but interactions with the tip and surface will unavoidably cause some distortions (generally a flattening and lateral spreading effect). This holds especially true for “soft” polymers such as PPO that have a low glass transition temperature (Tg = −70°C). Nevertheless, AFM visualization of DBC aggregates in fluid has generally yielded superior images and greater reproducibility than operation in air.

Amphiphilic DNA Block Copolymers

265

21. In general, this should be the same concentration of MgAc2 used to dilute the DBC solution. It will depend, however, on the material and may be much lower. In some cases, this step may be skipped entirely, in which instance the sample should be immediately applied. 22. The volume applied depends to some extent on the sample concentration; highly concentrated samples can be applied in a volume as low as 2 mL, though it is best to use a volume that completely covers the mica surface. 23. The total volume used in this rinsing step may vary with the strength of attachment of the DBC materials; typically, samples adsorbed in the presence of higher MgAc2 concentrations can survive more aggressive rinsing and drying and can thus have more background material removed. 24. SNL cantilevers have much sharper tips (nominal tip radius = 2 nm) than the standard NPS cantilevers. Thus, higher resolution imaging is possible, but additional care must be taken not to damage the sample, as these cantilevers exert much higher local pressures and penetrate more deeply into soft materials. 25. The relative spatial position of the folic acid on the micelle can be controlled by the design of the modified ODN utilized for hybridization. When applying 5¢-modified ODNs, the functionalities will be located at the rim of the micelle, while hybridization of 3¢-modified ODNs will result in micelles bearing the functional moiety close to the hydrophobic PPO core. 26. The coupling reaction of the amino-modified ODN and folic acid can be downscaled, but best results are obtained by employing reactants at high concentration. 27. The functionalization of the micelle corona by hybridization of complementary folic acid-functionalized ODNs can be performed using different DNA-b-PPO/ODN ratios, resulting in different functionalization densities. The DNA-b-PPO molecules (22-mer DNA, 6.8 kDa PPO) described here form micelles consisting of an average of 28 DNA-b-PPO molecules. Thus, complete functionalization (using equal amounts of DNA-b-PPO and folic acid-modified ODNs) yields micelles decorated with an average of 28 folic acid units (16).

Acknowledgments This work was supported by the EU (ERC starting grant, ECCell), the Nuffic (Huygens scholarship program), the DFG, and the Zernike Institute for Advanced Materials.

266

Zimmermann et al.

References 1. Lemaitre M., Bayard B., Lebleu B. (1987) Specific antiviral activity of a poly(L-lysine)conjugated oligodeoxyribonucleotide sequence complementary to vesicular stomatitis virus N protein mRNA initiation site. Proc. Natl. Acad. Sci. USA 84, 648–52. 2. Yang C. Y. J., Pinto M., Schanze K., Tan W. H. (2005) Direct synthesis of an oligonucleotidepoly(phenylene ethynylene) conjugate with a precise one-to-one molecular ratio. Angew. Chem. Int. Ed. 44, 2572–6. 3. Costioli M. D., Fisch I., Garret-Flaudy F., Hilbrig F., Freitag R. (2003) DNA purification by triple-helix affinity precipitation. Biotechnol. Bioeng. 81, 535–45. 4. Soh N., Umeno D., Tang Z. L., Murata M., Maeda M. (2002) Affinity precipitation separation of DNA binding protein using block conjugate composed of poly(N-isopropyla­crylamide) grafted double-stranded DNA and doublestranded DNA containing a target sequence. Anal. Sci. 18, 1295–9. 5. The Eyetec Study Group (2002) Preclinical and phase 1A clinical evaluation of an antiVEGF pegylated aptamer (EYE001) for the treatment of exudative age-related macular degeneration. Retina 22, 143–52. 6. Oishi M., Nagatsugi F., Sasaki S., Nagasaki Y., Kataoka K. (2005) Smart polyion complex micelles for targeted intracellular delivery of PEGylated antisense oligonucleotides containing acid-labile linkages. ChemBioChem 6, 718–25. 7. Jeong J. H., Kim S. W., Park T. G. (2003) A new antisense oligonucleotide delivery system based on self-assembled ODN-PEG hybrid conjugate micelles. J. Control. Release 183–91. 8. Kim S. H., Jeong J. H., Lee S. H., Kim S. W., Park T. G. (2008) Local and systemic delivery of VEGF siRNA using polyelectrolyte complex micelles for effective treatment of cancer. J. Control. Release 129, 107–16. 9. Alemdaroglu F. E., Herrmann A. (2007) DNA meets synthetic polymers – highly versatile hybrid materials. Org. Biomol. Chem. 5, 1311–20.

10. Alemdaroglu F. E., Ding K., Berger R., Herrmann A. (2006) DNA-templated synthesis in three dimensions: Introducing a micellar scaffold for organic reactions. Angew. Chem. Int. Ed. 45, 4206–10. 11. Safak M., Alemdaroglu F. E., Li Y., Ergen E., Herrmann A. (2007) Polymerase chain reaction as an efficient tool for the preparation of block copolymers. Adv. Mater. 19 1499–505. 12. Alemdaroglu F. E., Zhuang W., Zöphel L., et al. (2009) Generation of Multiblock copolymers by PCR: synthesis, visualization and nanomechanical properties. Nano Lett. 9, 3658–62. 13. Ding K., Alemdaroglu F. E., Börsch M., Berger R., Herrmann A. (2007) Engineering the structural properties of DNA block copolymer micelles by molecular recognition. Angew. Chem. Int. Ed. 46, 1172–5. 14. Alemdaroglu F. E., Wang J., Börsch M., Berger R., Herrmann A. (2008) Enzymatic control of the size of DNA block copolymer nanoparticles. Angew. Chem. Int. Ed. 47, 974–6. 15. Alemdaroglu F. E., Alemdaroglu C. N., Langguth P., Herrmann A. (2008) Shape dependent cellular uptake of dna nanoparticles. Macromol. Rapid. Commun. 29, 326–9. 16. Alemdaroglu F. E., Alemdaroglu C. N., Langguth P., Herrmann A. (2008) DNA Block copolymer micelles - A combinatorial tool for cancer nanotechnology. Adv. Mat. 20, 899–902. 17. The Sourcebook – A Handbook for Gel Electrophoresis, Cambrex Bio Science Rockland, Inc., Rockland, MA 18. Shuai X. T., Ai H., Nasongkla N., Kim S., Gao J. M. (2004) Micellar carriers based on block copolymers of poly(e-caprolactone) and poly(ethylene glycol) for doxorubicin delivery. J. Control. Release 98, 415–26. 19. Integrated DNA Technologies, Inc. webbased calculator for molar extinction coefficients of ODNs http://biophysics.idtdna. com/.

Part III Glycosyl and Lipid Conjugates

Chapter 16 Chemically Selective Liposome Surface Glyco-functionalization Hailong Zhang, Yong Ma, and Xue-Long Sun Abstract Liposome surface functionalization facilitates the enormous potential applications of liposomes, such as stabilizing and targeting carrier systems for delivering active molecules in biomedical research and applications. Cell surface carbohydrates have been an attractive model system for liposome surface functionalization for enhanced biomedical applications, such as site-specific and ligand-directed drug and gene delivery, multivalent inhibition, and vaccine adjuvant applications. The present protocol describes an efficient and chemically selective liposome surface glyco-functionalization method based on Staudinger ligation performed in phosphate-buffered saline buffer (pH 7.4) at room temperature. Specifically, a carbohydrate derivative carrying a spacer with an azide group is conjugated onto the surface of preformed liposomes carrying terminal triphenylphosphine groups. The compatibility of the reaction conditions for liposome surface functionalization was confirmed with dynamic light scattering and kinetic experiments monitoring the leakage of entrapped 5,6-carboxyfluorescein. Key words: Liposome, Surface glyco-functionalization, Carbohydrate, Staudinger ligation, Drug targeting, Carrier system

1. Introduction Liposomes, comprised of (phospho)lipid molecules that selfassemble into spherical, self-closed structures, have been extensively studied as models of cell membrane structure and mostly as carriers for drug/gene delivery applications (1, 2). Presently, studies of liposome surface functionalization are facilitating the enormous potential application of liposomes in several areas of research (3, 4). Cell surface carbohydrates have been a particularly attractive model for liposome surface modification due to their protein-rejecting ability, biodegradability, low toxicity, and especially their cell targeting ability through specific binding Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_16, © Springer Science+Business Media, LLC 2011

269

270

Zhang, Ma, and Sun

interactions with receptors expressed on the surface of targeted cells (5). For example, monosialoganglioside (GM1) enhances the circulation lifetime of liposomes at a level comparable to that observed for polyethylene glycol (PEG) (6). Furthermore, sialyl LewisX-decorated liposomes have been demonstrated to target the delivery of drugs to endothelial cells based on the site-specific expression of E- and P-selectin in blood vessels during inflammation (7). Carbohydrate-decorated liposomes have also been used as multivalent platforms to inhibit carbohydrate-mediated cell adhesion. For example, sialic acid-decorated liposomes showed strong inhibitory activity against influenza hemagglutinin and neuraminidase (8). Moreover, carbohydrate-anchored liposomes have been used as vaccine adjuvants. For example, oligomannosecoated liposomes have been shown to induce a T helper-type 1 immune response and to control the early stages of African trypanosomiasis (9). Taken together, carbohydrate-decorated liposomes have paved the way for creating biostable, multivalent, site-specific, and ligand-directed drug and gene delivery systems that are highly desirable for potential use in therapeutic and immunological applications. Various techniques have been developed for liposome surface functionalization. Conventional methods to prepare surfacefunctionalized liposomes involve the initial synthesis of the key lipid–ligand conjugate, followed by the formulation of the liposome with all the lipid components. In this direct liposome formation method, some of the valuable ligands inevitably become oriented toward the enclosed aqueous compartment and thus become unavailable for their intended interaction with their target molecules. Such a synthetic approach is particularly undesirable if the targeting ligand is only available in minimum amounts. Furthermore, lipid–ligand conjugates typically have limited solubility and stability in solvents or they are incompatible with other reagents at various stages of manufacture. Recently, alternative chemical modification methods, which in most cases involve the coupling of biomolecules to the surface of preformed vesicles that carry functionalized (phospho)lipid anchors, have been developed (10, 11). To date, variable levels of success using approaches based on amide coupling (12), thiol-maleimide coupling (13), imine chemistry (14), hydrazone linkage (15), and “click” chemistry (16) have been described. However, in many cases, there is a lack of specificity resulting in the formation of an uncontrolled number of covalent bonds between the liposome and biomolecules of interest, which may in turn diminish the bioactivity of the attached biomolecule and the stability of the liposome. Copper(I)catalyzed [3 + 2] cycloaddition (termed “click” chemistry) (17), which can occur efficiently in aqueous media at room temperature and selectively between azide and alkyne, has been investigated as a novel generic chemical tool for the facile in situ surface

Chemically Selective Liposome Surface Glyco-functionalization

271

modification of liposomes. However, a key limitation of “click” chemistry is the required use of a Cu(I) catalyst, which results in residual copper being present in the modified liposomes and which could be a potential concern for their biological application downstream. Staudinger ligation, in which an azide and triphenylphosphine selectively react to form an amide bond, has been used for the chemically selective and biocompatible modification of recombinant proteins (18) and living cells (19) under native conditions. Significantly, the reaction can occur in high yields at room temperature in aqueous solvents and without any catalyst, and is compatible with the unprotected functional groups of a wide range of biomolecules. Herein, an efficient and chemically selective liposome surface functionalization protocol employing lactose as a model carbohydrate was developed based on Staudinger ligation (Fig. 2) (20). The high specificity and high yield, as well as the biocompatible nature of the reaction conditions, make the Staudinger ligation approach an attractive alternative to most other protocols currently available for liposome surface functionalization.

2. Materials 2.1. Synthesis of Anchor Lipid DPPETriphenylphosphine 3

1. 1,2-Dipalmitoyl-sn-glycero-3-phosphoethanolamine (DPPE) (Avanti Polar Lipids, Inc., Alabaster, AL). 2. Triethylamine (Et3N). 3. Succinimidyl 3-diphenylphosphino-4-methoxycarbonylbenzoate was synthesized as described in ref. 21. 4. Silica gel (230–400 mesh) (Sigma, St. Louis, MO). 5. Silica gel aluminum-backed thin layer chromatography (TLC) plates (250 mm thickness) (Whatman, Piscataway, NJ). 6. Dichloromethane (Anhydrous). 7. Chloroform (CHCl3) (ACS reagent grade). 8. Methanol (ACS reagent grade). 9. Concentrated H2SO4. 10. Deuterochloroform (99.8 at.% D). 11. 10% (v/v) H2SO4 in methanol: Dissolve 10 mL of concentrated H2SO4 into 90 mL of methanol slowly.

2.2. Preparation of Liposomes

1. 1,2-Dipalmitoyl-sn-glycero-3-phosphocholine (DPPC) (Avanti Polar Lipids, Inc.). 2. Cholesterol.

272

Zhang, Ma, and Sun

3. DPPE-triphenylphosphine 3 was synthesized as described in Subheading 3.1. 4. Phosphate-buffered saline (PBS) 10× stock: 1.37  M NaCl, 27 mM KCl, 100 mM Na2HPO4, 18 mM KH2PO4 (if necessary, adjust to pH 7.4 with HCl). Prepare a 1× working solution by dilution of one part of 10× PBS stock with nine parts deionized water. 5. Polycarbonate membranes (pore size: 2,000, 600, 200, 100 nm) (Whatman). 6. 5,6-Carboxyfluorescein (Sigma). 7. Sephadex® G-50 (Sigma). 2.3. Liposome Surface Glycofunctionalization

1. 2-Azidoethyl lactoside was synthesized using literature method (22). 2. Deionized water with a resistivity of 18.2 MW cm was used as a solvent in all reactions. 3. 0.5% (v/v) Triton X-100 solution: Dissolve 0.5 mL of Triton X-100 (Sigma) into 99.5 mL of deionized water. 4. 0.5% (w/v) Phenol solution: Dissolve 0.5 g of phenol (Sigma) into 100 mL of deionized water. 5. Lectin (from Arachis hypogaea, 120 kDa) (Sigma).

2.4. Analytical Instrumentation

1. 1H NMR, 13C NMR, and 31P NMR spectra were recorded with a Varian INOVA 300 MHz spectrometer (see Note 1). 2. Liposomes were prepared by using a LIPEX™ Extruder (Northern Lipids, Inc, Canada). 3. Dynamic light scattering (DLS) measurements were recorded with a 90Plus particle size analyzer (Brookhaven Instruments Corporation, Holtsville, NY). 4. Fluorescence spectra were measured with a FluoroMax-2 spectrofluorometer (ISA, Inc, Edison, NY). 5. Electrospray ionization mass spectrometry (ESI-MS) was conducted with an ion trap mass spectrometer (Esquire HCT, Bruker Daltonics Inc.). 6. UV absorption spectra were measured with a Cary 50 Bio UV-visible spectrophotometer (Varian, Palo Alto, CA).

3. Methods 3.1. Synthesis of Anchor Lipid DPPETriphenylphosphine 3 (see Note 2)

1. DPPE (0.26 g, 0.376 mmol) was dissolved in 40 mL of anhydrous CH2Cl2 in a 100 mL round-bottom flask. 2. 0.8 mL of Et3N was added to the above solution.

Chemically Selective Liposome Surface Glyco-functionalization

273

3. After stirring for 30  min at room temperature (RT), a solution of succinimidyl 3-diphenylphosphino-4-methoxycarbonylbenzoate 1 (0.23 g, 0.500 mmol) in 10 mL of anhydrous CH2Cl2 was added. 4. The reaction mixture was stirred for 24  h at RT under an argon atmosphere. 5. The reaction mixture was concentrated under vacuum evaporation to afford crude product lipid DPPEtriphenylphosphine 3. 3.2. Purification of DPPETriphenylphosphine 3 by Silica Gel Chromatography

1. The crude product lipid DPPE-triphenylphosphine 3 was dissolved into 3 mL of CHCl3 and then the solution was loaded onto a pre-packed silica gel column (2.0 × 20 cm). 2. The column was eluted with chloroform/methanol (4:1, v/v), and fractions of 10 mL each were collected. 3. All fractions were checked by TLC (developing with chloroform/methanol (4:1, V/V), charring the TLC sample plate on a hot plate after dipping in 10% (v/v) H2SO4 in methanol). 4. Fractions containing the desired compound (Rf value = 0.35) were combined and concentrated under vacuum evaporation to afford pure lipid DPPE-triphenylphosphine 3 (0.283  g, 73% yield). 5. 1H NMR, 13C NMR, and 31P NMR spectra were recorded with a Varian INOVA 300  MHz spectrometer. In all cases, the sample concentration was 10  mg/mL in CDCl3 (see Note 3).

3.3. Preparation of TriphenylphosphineContaining Liposomes

1. DPPC (30 mg, 40.87 mmol), cholesterol (8 mg, 20.4 mmol), and DPPE-triphenylphosphine 3 (3.5  mg, 3.2  mmol) were dissolved in 3.0 mL of chloroform in a 100-mL round-bottom flask. 2. The solvent was gently removed on an evaporator under reduced pressure to form a thin lipid film on the flask wall and kept in a vacuum chamber overnight. 3. The lipid film was swelled by adding 2.5 mL of PBS buffer (pH 7.4) to form a lipid suspension in the round-bottom flask. 4. The lipid suspension contained inside the round-bottom flask was subjected to ten freeze–thaw cycles in liquid N2 and then immersed in a 65°C water bath. 5. The lipid suspension thus formed was extruded repeatedly ten times by gradually passing through polycarbonate membranes with pore sizes of 2,000, 600, 200, and 100 nm using a LIPEX™ Extruder (see Note 4) to afford the small unilamellar vesicles (SUVs) (see Note 5).

274

Zhang, Ma, and Sun

3.4. Preparation of TriphenylphosphineContaining Liposomes Encapsulating 5,6Carboxyfluorescein

1. DPPC (30 mg, 40.87 mmol), cholesterol (8 mg, 20.4 mmol), and DPPE-triphenylphosphine 3 (3.5  mg, 3.2  mmol) were dissolved in 3.0 mL of chloroform in a 30-mL round-bottom flask. 2. The solvent was gently removed on an evaporator under reduced pressure to form a thin lipid film on the flask wall and kept in a vacuum chamber overnight. 3. The lipid film was swelled by adding 2.5 mL of PBS buffer (pH 7.4) containing 85 mM carboxyfluorescein (CF) to form a lipid suspension in the round-bottom flask. 4. The lipid suspension contained inside the round-bottom flask was subjected to ten freeze–thaw cycles in liquid N2 and then immersed in a 65°C water bath. 5. The lipid suspension thus formed was extruded repeatedly ten times by gradually passing through polycarbonate membranes with pore sizes of 2,000, 600, 200, and 100 nm using a LIPEX™ Extruder to afford the SUVs. 6. The PBS solution containing the SUVs was passed through a Sephadex® G-50 column (1.5 × 20  cm) to remove unencapsulated CF from the CF encapsulated within the liposomes.

3.5. Liposome Surface Glycofunctionalization

1. 2-Azidoethyl lactoside (4  mg, 0.01  mmol) was dissolved in 0.2 mL of PBS (pH 7.4) buffer (pretreated with argon bubbling prior to use) and added into 2 mL of the PBS (pH 7.4)/liposome solution obtained in Subheading 3.3 above (see Note 6). 2. The reaction was gently shaken at RT for 6 h under an argon atmosphere. 3. Unreacted 2-azidoethyl lactoside was removed by gel filtration by passing through a Sephadex® G-50 column (1.5 × 20 cm) to afford the lactosylated liposome product. 4. The size of the liposomes during the Staudinger ligation reaction was monitored by using a 90Plus particle size analyzer (see Note 7). 5. A control experiment was also conducted in the absence of 2-azidoethyl lactoside. 6. The surface glyco-functionalization of liposomes encapsulating 5,6-CF was conducted by using the liposome solution obtained in Subheading  3.4 and by following steps 1–4 above.

3.6. Fluorescent Leakage Determination (see Note 8)

1. During the liposome surface glyco-functionalization reaction (Subheading  3.5), a 20-mL sample of the reaction solution was taken at 0, 1, 2, 3, 4, 5, and 6 h. The 20-mL sample was mixed with 1,980 mL PBS (pH 7.4) buffer, and the fluorescence

Chemically Selective Liposome Surface Glyco-functionalization

275

intensity of the diluted solution was measured with a FluoroMax-2 spectrofluorometer. 2. Subsequently, 20 mL of 0.5% (v/v) Triton X-100 was added into the above solution to destroy the liposomes. Following this, 20  mL of solution containing the destroyed liposomes was taken and further diluted ten times with PBS (pH 7.4) buffer. The fluorescence intensity of the diluted solution was measured with a FluoroMax-2 spectrofluorometer. 3. Fluorescence measurements were also conducted with control (nonlactosylated) liposomes obtained from Subheading  3.5, step 5 that were not reacted with 2-azidoethyl lactoside. 3.7. Determination of the Concentration of Lactose on the Liposome Surface

1. 0.5 mL of 0.5% (w/v) phenol solution was added into 0.5 mL of lactosylated liposome and mixed well. 2. 2.5 mL of concentrated H2SO4 was added directly into the solution. 3. The mixture was vortexed, and allowed to stand for 30 min at room temperature. 4. The optical absorbance reading of the solution above was taken at 490 nm with a Cary 50 Bio UV-visible spectrometer. 5. The amount of lactose grafted onto the surface of the liposomes was calculated from standard calibration curves prepared as described by Saha et al. (23) using deionized water solutions containing free lactose.

3.8. Characterization of Specific Binding Between Lactose on the Liposome Surface and Lectins (see Note 9)

3.9. Characterization of Stability of Liposome with DLS (see Note 10)

1. 0.5 mL of PBS (pH 7.4) solution containing 1 mg/mL lectin (A. hypogaea, 120 kDa) was added to 0.5 mL of PBS (pH 7.4) solution containing lactosylated liposomes. 2. The size of the liposomes was monitored with DLS over time. 3. A control experiment was also conducted with nonlactosylated liposomes. 1. The size of the lactosylated liposomes in PBS (pH 7.4) buffer at RT was monitored with DLS over time. 2. The size of the control liposomes without lactose in PBS (pH 7.4) buffer at RT was monitored with DLS over time.

4. Notes 1. In all cases, the sample concentration was 10  mg/mL, and the appropriate deuterated solvent was used as an internal standard.

276

Zhang, Ma, and Sun

Fig. 1. Scheme illustrating the synthesis of DPPE-triphenylphosphine conjugate 3.

Fig. 2. 1H NMR (a), 13C NMR (b), and 31P NMR (c) spectrum of DPPE-triphenylphosphine conjugate 3 (in CDCl3).

2. The terminal triphenylphosphine-carrying anchor lipid 3 was synthesized by amidation of commercially available DPPE with 3-diphenylphosphino-4-methoxycarbonylbenzoic acid NHS active ester 1 (Fig. 1). 3. Anchor lipid 3 was characterized by 1H NMR (Fig. 2a), 13C NMR (Fig. 2b), and 31P NMR (Fig. 2c) spectra recorded with a INOVA 300 MHz spectrometer. A typical chemical shift at −3.74 ppm in the 31P NMR spectrum confirmed the presence of the phosphine group in anchor lipid 3. 4. SUVs composed of saturated phospholipids (DPPC) and cholesterol (2:1 molar ratio) and 5 mol% of the anchor lipid 3 were prepared by sequential extrusion through polycarbonate membranes with pore sizes of 2,000, 600, 200 and 100  nm at 65°C. The extrusion was conducted by using a LIPEX™ Extruder following the manufacturer’s Extruder Assembly and Operation Manual. 5. This extrusion process predominately produced SUVs with an average mean diameter of 120 ± 5 nm, as determined by DLS (Fig. 4a). 6. Conjugation of 2-azidoethyl lactoside to the preformed triphenylphosphine-anchored liposomes was performed in PBS buffer (pH 7.4) at RT under an argon atmosphere for 6  h (Fig.  3). An argon atmosphere was undertaken to prevent oxidation side reactions of the triphenylphosphine on the liposome surface. 7. DLS was used to verify the integrity of the vesicles during and after the coupling reaction. As shown in Fig. 4, there is no

Chemically Selective Liposome Surface Glyco-functionalization

277

Fig.  3. Schematic illustration of liposome formation and liposome surface glyco-functionalization through Staudinger ligation.

Fig.  4. Monitoring of liposome particle size by dynamic light scattering (DLS). (a) Before conjugation and (b) after conjugation.

significant change in the size of the vesicles during and after the conjugation reaction. Therefore, the reaction conditions described above do not alter the integrity of the liposomes. 8. Fluorescent leakage from liposomes having encapsulated, self-quenching concentrations of 5,6-carboxyfluorescein was investigated to test whether the conjugation condition could provoke some leakage of the liposomes (24). As shown in the results, there was no apparent leakage triggered by the conjugation reaction (Fig.  5a) when compared with liposomes incubated in the absence of the coupling reagents (Fig. 5b). Finally, 20 mL of 0.5% (v/v) Triton X-100 was added into the reaction solution to destroy the liposomes; a significant amount of fluorescent incensement in the reaction solution and the control solution confirmed the integrity of the liposomes during the reaction. 9. To determine whether the grafted lactose residues are easily accessible at the surface of the liposomes, a lectin-binding assay was conducted by incubating the lactosylated liposomes in the presence of b-galactose-specific binding lectin (from A. hypogaea) in PBS (pH 7.4) buffer. After 30 min, apparent visible aggregation formed as monitored by DLS (Fig. 6a).

278

Zhang, Ma, and Sun

Fig. 5. Kinetics of carboxyfluorescein release from liposomes reacted with 2-azideethyl lactoside (a) and in the absence of 2-azideethyl lactoside (b, control experiment).

Fig. 6. DLS monitoring of aggregation of lactosylated liposomes in the presence of lectin (a) and without lectin (b).

Fig. 7. Stability of liposomes without lactose (a) and stability of lactosylated liposomes (b) as monitored by DLS.

Chemically Selective Liposome Surface Glyco-functionalization

279

In contrast, neither aggregation nor any size change was observed with liposomes alone (Fig. 6b), confirming that the agglutination was due to a specific recognition of the lactose residues on the surface of the liposomes by the lectin via multivalent interactions. 10. The stability of the lactosylated liposomes was evaluated by comparison with the stability of liposomes without lactose at RT as monitored by DLS. As shown in Fig. 7, both types of liposomes showed good stability during the 8-day monitoring period. However, the liposomes without lactose began to collapse and aggregate after the 9th day (Fig. 7a), while there was no apparent size change for the lactosylated liposomes until the 15th day (Fig. 7b). This result demonstrates that the presence of lactose on the liposome surface provides a steric barrier that prevents liposome aggregation.

Acknowledgments The authors acknowledge financial support under grants from the American Health Assistance Foundation (AHAF-H2007027), from NIH-NHLBI (1R01HL102604-01), and from the Startup Fund from Cleveland State University. Thanks to Dr. Dale Ray at CCSB for the NMR studies and Dr. Xiang Zhou at CSU for the mass spectrometry studies. References 1. Philippot, J. R. and Schuber F. (1995) Liposomes as Tools in Basic Research and Industry, CRC Press, Boca Raton. 2. Allen, T. M. and Cullis, P. R. (2004) Drug delivery systems: Entering the mainstream. Science 303, 1818–1822. 3. Torchilin, V. P. (2005) Recent advances with liposomes as pharmaceutical carriers. Nat. Rev. Drug Discov. 4, 145–160. 4. Allen, T. M., Hansen C., Martin F., Redemann C., and Young, A.Y. (1991) Liposomes containing synthetic lipid derivatives of poly ethylene glycol show prolonged circulation half-lives in vivo. Biochim. Biophys. Acta 1066, 29–36. 5. Sihorkar, V., Vyas, S. P. (2001) Potential of polysaccharide anchored liposome in drug delivery, targeting and immunization. J. Pharmceut. Sci. 4, 138–158. 6. Mehvar, R. (2003) Recent Trends in the use of polysaccharides for improved delivery of therapeutic agents: Pharmacokinetic and pharmaco-

dynamic perspectives. Curr. Pharm. Biotechnol. 4, 283–302. 7. Stahn, R., Grittner, C., Zeisig, R., Karsten, U., Wenzel, F. (2001) Sialyl Lewis(x)-liposomes as vehicles for site-directed, E-selectin-mediated drug transfer into activated endothelial cells. Cell. Mol. Life Sci. 58, 141–147. 8. Sun, X. L., Kanie Y., Guo, C. T., Kanie O., Suzuki, Y. and Wong, C. H. (2000) Syntheses of C-3 modified sialylglycosides as selective inhibitors of influenza hemagglutinin and neuraminidase. Eur. J. Org. Chem. 14, 2643–2653. 9. Kuboki, N., Yokoyama, N., Namangala, B., Okamura, M., Inoue, N., Takagi, H., Nakayama, T., Nishikawa, Y., Ikehara, Y., Kojima, N. (2008) Adjuvant effect of oligomannose-coated liposome-based platform for vaccine against African trypanosomosis. J. Protozool. Res. 18, 1–10. 10. Sapra, P. and Allen, T. M. (2003) Ligandtargeted liposomal anticancer drugs. Prog. Lipid Res. 42, 439–462.

280

Zhang, Ma, and Sun

11. Nobs, L., Buchegger, F., Gurny, R. and Allemann E. (2004) Current methods for attaching targeting ligands to liposomes and nanoparticles. J. Pharm. Sci. 93, 1980–1982. 12. Kung, V. T. and Redemann, C. T. (1986) Synthesis of carboxyacyl derivatives of phosphatidylethanolamine and use as an efficient method for conjugation of protein to liposomes. Biochim. Biophys. Acta 862, 435–439. 13. Schelte, P., Boeckler, C., Frisch, B. and Schuber, F. (2000) Differential reactivity of maleimide and bromoacetyl functions with thiols: application to the preparation of liposomal diepitope constructs. Bioconjug. Chem. 11, 118–123. 14. Nakano, Y., Mori, M., Nishinohara, S., Takita, Y., Naito, S., Kato, H., Taneichi, M., Komuro, K., Uchoda, T. (2001) Surface-linked liposomal antigen induces IgE-selective unresponsiveness regardless of the lipid components of liposomes. Bioconjug. Chem. 12, 391–395. 15. Bourel-Bonnet, L., Pecheur, E. I., Grandjean, C., Blanpain, A., Baust, T., Melnyk, O., Hoflack, B., Gras-Masse, H. (2005) Anchorage of synthetic peptides onto liposomes via hydrazone and alpha-oxo hydrazone bonds. preliminary functional investigations. Bioconjug. Chem. 16, 450–457. 16. Hassane, F. S., Frisch, B., Schuber, F. (2006) Targeted liposomes: convenient coupling of ligands to preformed vesicles using “click chemistry”. Bioconjug. Chem. 17, 849–854. 17. Sun, X. L., Stabler, C, Cazalis, C., Chaikof, E. L. (2006) Carbohydrates and protein

immobilization onto solid surface by sequential Diels-Alder and azide-alkyne cycloaddition. Bioconjug. Chem. 17, 52–57. 18. Kiick, K. L.; Saxon, E.; Tirrell, D. A.; Bertozzi, C. R. (2002) Incorporation of azides into recombinant proteins for chemoselective modification by the Staudinger ligation. Proc. Natl. Acad. Sci. USA 99, 19–24. 19. Prescher, J. A., Dube, D. H., Bertozzi, C. R. (2004) Chemical remodeling of cell surfaces in living animals. Nature 430, 873–877. 20. Zhang, H., Ma, Y., Sun, X.-L. (2009) Chemically-selective surface glyco-functionalization of liposomes through Staudinger ligation. Chem. Commun. 3032–3034. 21. Seo, T. S., Wang, C., Li, Z. M., Ruparel, H., Ju. J. Y. (2003) Site-specific fluorescent labeling of DNA using Staudinger ligation. Bioconjug. Chem. 14, 697–701. 22. Sun, X. L., Grande, D., Baskaran, S., and Chaikof, E. L. (2002) Glycosaminoglycanmimetic biomaterials 4: Synthesis of sulfated lactose-based glycopolymers that exhibit anticoagulant activity. Biomacromolecules 3, 1065–1070. 23. Saha, S.K., Brewer, C.F. (1994) Determination of the concentrations of oligosaccharides, complex type carbohydrates, and glycoproteins using the phenol-sulfuric acid method. Carbohydr. Res. 254, 157–167. 24. Barbet, J., Machy, P., Truneh, A. and Leserman, L. D. (1984) Weak acid-induced release of liposome-encapsulated carboxyfluorescein. Biochim. Biophys. Acta 772, 347–356.

Chapter 17 Bioconjugation Using Mutant Glycosyltransferases for the Site-Specific Labeling of Biomolecules with Sugars Carrying Chemical Handles Boopathy Ramakrishnan, Elizabeth Boeggeman, Marta Pasek, and Pradman K. Qasba Abstract This chapter presents a technique that employs mutant glycosyltransferase enzymes for the site-specific bioconjugation of biomolecules via a glycan moiety to facilitate the development of a targeted drug delivery system. The target specificity of this methodology is based on unique sugar residues that are present on glycoproteins or engineered glycopeptides. The glycosyltransferases used in this approach have been manipulated in a way that confers the ability to transfer a modified sugar residue with a chemical handle to a sugar moiety of the glycoprotein or to a polypeptide tag of an engineered nonglycoprotein. The availability of the modified sugar moiety thus makes it possible to link cargo molecules at specific sites. The cargo may be comprised of, for example, biotin or fluorescent tags for detection, imaging agents for magnetic resonance imaging (MRI), or cytotoxic drugs for cancer therapy. Key words: Glycosyltransferase, Monoclonal antibody (mAb), Asn-linked glycan (N-glycan), Single-chain antibody (scFv), Human polypeptidyl-a-N-acetylgalactosaminyltransferase (h-ppaGalNAc-T), a1,3-Galactosyltransferase (a3Gal-T), 2-Acetonyl-2-deoxy-galactose (C2-keto-Gal)

1. Introduction Recently, novel glycosyltransferases have been designed with broader donor specificities that can transfer a sugar residue with a chemically reactive functional group to a specific terminal moiety of a glycoconjugate (1–3). Site-specific labeling of monoclonal and single-chain antibodies and other therapeutic proteins with bioactive molecules are of great interest in the detection and treatment of cancer. Currently, several chemical approaches are being used for the conjugation of biomolecules through free

Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_17, © Springer Science+Business Media, LLC 2011

281

282

Ramakrishnan et al.

cysteine or lysine residues present on the surface of proteins (4). However, conjugation through cysteine residues requires special handling of proteins (5), whereas conjugation through lysine residues results in the random conjugation of bioactive molecules (6). We have recently developed a chemoenzymatic approach to this problem by utilizing a unique sugar moiety that can be transferred by a glycosyltransferase to a glycoprotein or to the peptide tag of a nonglycoprotein (7–10). 1.1. Site-Specific Labeling of Monoclonal Antibodies Through Their N-Glycans

Within the Fc region of a monoclonal immunoglobulin G molecule (IgG), each heavy chain carries a single Asn-linked glycan chain (N-glycan) at its CH2 domain. Since the N-glycan is located far away from the Fab region (the antigen binding region of the IgG), any modification of this glycan is not expected to affect the antigen binding properties of the antibody; therefore, we have chosen to utilize this site for site-specific conjugation (9). Although N-glycans have previously been utilized for conjugation by others through aldehyde-based chemistry approaches, the process of creating an aldehyde moiety in these N-glycans is rather harsh and can be destructive to the native protein itself (11). In contrast, our approach to address this problem employs a chemoenzymatic-based method, which is relatively mild (9). In human IgG1, the Fc N-glycans are complex biantennary structures with variable galactosylation – either 0, 1, or 2 terminal galactoses may be present, corresponding to the G0, G1, and G2 glycoforms, respectively; and only <10–14% are sialylated (12). The enzyme b-4galactosyltransferase 1 (b4Gal-T1) is responsible for the presence of the galactose (Gal) moieties in these N-glycans, where it transfers Gal from uridine 5¢-diphosphate galactose (UDP-UDP-Gal) to the acceptor sugar N-acetylglucosamine (GlcNAc) present at the nonreducing end of N-glycans. Based on structure–function studies, we have engineered this enzyme to transfer not only galactose, but also N-acetylgalactosamine (GalNAc) with equal efficiency (1). We find that this mutant enzyme, known as Y289L-b4Gal-T1, can also transfer modified galactose sugars with a chemical handle at the second carbon position, such as 2-acetonyl-2-deoxy-galactose (C2-keto-Gal) or N-acetyl-azido-galactosamine (GalNAz), from their respective UDP-sugar donor substrates (2, 7). Upon such transfer, the unique chemical handle can be used to conjugate bioactive molecules with orthogonal reactive groups (7, 9).

1.2. Site-Specific Labeling of SingleChain Antibodies

Single-chain antibodies – rather than their full-length IgG counterparts – are increasingly being used for immunotherapy since they are easily expressed in large amounts in Escherichia coli as soluble proteins. In order to label these molecules in a site-specific manner, we have developed a novel chemoenzymatic method using the human polypeptidyl-a-N-acetylgalactosaminyltransferase

Bioconjugation Using Mutant Glycosyltransferases

283

(h-ppaGalNAc-T) enzyme. Generally, ppaGalNAc-T transfers GalNAc from UDP-GalNAc to the side chain hydroxyl group of Ser or Thr residues present in a linear acceptor peptide that is at least 13 amino acids (AA) long. Moreover, it has also been shown that ppaGalNAc-T can transfer GalNAz from UDP-GalNAz to Ser or Thr residues on the acceptor peptide (8, 13). We recently engineered the specific acceptor peptide substrate for the h-ppaGalNAc-T2 enzyme (a homologous member of the ppGalNAc-T family of enzymes in humans) as a fusion peptide at the C-terminus of the anti-HER2 scFv molecule (8, 10). The resulting fusion peptide was used as an acceptor substrate for glycosylation with C2-keto-Gal or GalNAz using ppa-GalNAc-T2; and this was then followed by conjugation of the bioactive scFv fusion molecule with a fluorescent probe (e.g., Alexa Fluor® 488 dye) containing an orthogonal reactive group (10). 1.3. Chemoenzymatic Labeling Other Glycoproteins via Their LacNAc Moiety

Results from our laboratory and others have shown that a limited number of residues in the sugar donor binding pocket of glycosyltransferases determine the specificity of these enzymes, and that mutation of only a few residues can alter their sugar donor specificity (1, 3). To detect the most prevalent sugar moiety found on the surface of cells, viz., the terminal N-acetyllactosamine (LacNAc (Galb1-4GlcNAc)), the bovine enzyme a1,3-galactosyltransferase (a3Gal-T) was engineered to produce a mutant 1,3-galactosaminyltransferase (a3GalNAc-T) enzyme (3). The wild-type a3Gal-T enzyme normally transfers galactose from UDP-Gal to the terminal galactose sugar of the LacNAc or lactose disaccharide moiety, forming a product with a Gala1-3-linkage. We have rationally mutated the sugar donor binding site residues of a3Gal-T (through site-directed mutagenesis of His280, Ala281, and Ala282 (280HAA282) to Ser280, Gly281, and Gly282 (280SGG282-a3Gal-T), respectively), altering its sugar donor specificity from Gal to GalNAc. Our results showed that the 280SGG282 mutant enzyme also selectively transfers the modified sugar C2-keto-Gal from its UDP derivative to free LacNAc residues present at the nonreducing end of glycans of an asialoglycoprotein, asialofetuin. The chemical handle at the C2 position of the modified galactose was further used in biotinylation and subsequent detection of the terminal LacNAc moiety in glycoproteins by chemiluminescence methods.

2. Materials The recombinant glycosyltransferases, Y289L-b4Gal-T1, SGG282-a3Gal-T, and h-ppaGalNAc-T2, can be prepared according to procedures described previously (14). The enzyme 280

284

Ramakrishnan et al.

Y289L-b4Gal-T1 can also be obtained commercially from Invitrogen (Eugene, OR). The galactose analogue uridine 5¢-diphospho-2-acetonyl-2-deoxy-galactose (UDP-C2-keto-Gal) can be synthesized as described previously (2). Alternatively, uridine 5¢-diphospho-N-acetyl-azido-galactosamine (UDP-GalNAz) can also be obtained commercially from Invitrogen. 2.1. Degalactosylation of Monoclonal Antibodies

1. Recombinant monoclonal antibodies (mAbs): Rituxan (Rituximab), Remicade (Infliximab), Avastin (Bevacizumab), and Herceptin (Trastuzumab) (Genentech, Inc., South San Francisco, CA). 2. Washing buffer: 50  mM Sodium phosphate, pH 6.0 (see Note 1). 3. Microcon Ultracel YM-50 centrifugal filter devices (Millipore Corporation, Bedford, MA). 4. Recombinant Streptococcus pneumoniae b1,4-galactosidase (Calbiochem, San Diego, CA).

2.2. Protein A Affinity Chromatography of Degalactosylated mAbs

1. Protein A-Sepharose® 4B columns (Invitrogen). 2. Binding buffer: 1× Phosphate-buffered saline (PBS), pH 7.4. 3. Washing buffer: 1× PBS, pH 7.4. 4. Elution buffer: 100 mM Glycine–HCl, pH 2.7. 5. Neutralizing buffer: 1 M Tris–HCl, pH 8.0. 6. Microcon Ultracel (Millipore).

YM-50

centrifugal

filter

devices

7. Bio-Rad Protein Assay Kit (Bio-Rad, Hercules, CA). 2.3. Mass Spectrometry Analysis of Oligosaccharides Released After PNGase F Treatment of Degalactosylated mAbs

1. Peptide N-glycosidase F (PNGase F) (New England Biolabs, Ipswich, MA). 2. 10× G7 reaction buffer (supplied with the PNGase F enzyme) (New England Biolabs). 3. Microspin active charcoal columns (Harvard Apparatus, MA). 4. 30% (v/v) acetonitrile in water. 5. Matrix-assisted laser desorption/ionization-time of flight (MALDI-TOF) mass spectrometer.

2.4. Transfer of C2-Keto-Galactose to Degalactosylated mAbs or to Asialofetuin Using Engineered Glycosyltransferases

1. 10 mM UDP-C2-keto-Gal (2). 2. Degalactosylated mAbs (prepared as described in Subheading 3.1). 3. Asialofetuin (Sigma–Aldrich, St. Louis, MO). 4. 500 mM Tris–HCl, pH 8.0. 5. 100 mM MnCl2.

Bioconjugation Using Mutant Glycosyltransferases

2.5. Biotinylation of C2-Keto-GalactoseModified mAbs or Asialofetuin

1. 3 M sodium acetate (NaOAc) buffer, pH 3.9.

2.6. SDS-PAGE/ Western Blotting Analysis of AsialoAgalacto-mAbs or Asialofetuin

1. PowerEase® 500 electrophoresis power supply.

285

2. N ¢-Aminooxymethylcarbonylhydrazino-D-biotin (AOB) (Dojindo Laboratories, Japan). For biotinylation reactions, prepare a fresh stock solution containing 30  mM AOB in water.

2. Tris–glycine precast gels, 14% (w/v) gel (Invitrogen). 3. Tris–glycine SDS running buffer (10×) (Invitrogen). 4. Nitrocellulose membrane (0.45 mm pore size) (Invitrogen). 5. Tris–glycine transfer buffer (25×) (Invitrogen). 6. Blocking and washing solutions: 5% (w/v) dry milk in 0.02% (v/v) Tween-20. 7. Probe solution: Streptavidin-conjugated horseradish peroxidase (HRP) (GE Healthcare, Piscataway, NJ) diluted 1:4,000 (v/v) in 1× PBS (pH 7.4), 0.02% (v/v) Tween-20 containing 3% (w/v) bovine serum albumin Fraction V (BSA) (Roche Diagnostics, Mannheim, Germany). 8. ECL™ Chemiluminescence Detection Reagents 1 and 2 (GE Healthcare). 9. Kodak® Biomax™ XAR film.

2.7. Fluorescent Labeling of C2-KetoGal-Modified mAbs

1. AlexaFluor® 488C5-aminooxyacetamide,bis(triethylammonium) salt, 1 mg/mL in dimethyl sulfoxide (DMSO) (Invitrogen). 2. 3 M Sodium acetate buffer, pH 4.9. 3. Micro Bio-Spin 30 size exclusion chromatography spin columns prepacked with Bio-Gel P-30 (Bio-Rad). 4. 10× PBS, pH 7.4. 5. FMBIO II fluorescence imaging system (Hitachi).

2.8. Glycosylation of Anti-HER2 scFv Fusion Protein Using an Engineered Glycosyltransferase

1. Anti-HER2 scFv fusion protein containing one or multiple O-glycosylation sites, prepared as described previously (10). 2. h-ppaGalNAc-T2 enzyme, prepared as described previously (14). 3. 10 mM UDP-C2-keto-Gal (2). 4. 250 mM Tris–HCl, pH 8.0. 5. 100 mM MnCl2.

2.9. Purification of Glycosylated Anti-HER2 scFv Fusion Protein

1. Ammonium sulfate. 2. Microcon Ultracel YM-10 centrifugal filter devices (Millipore).

286

Ramakrishnan et al.

2.10. Fluorescent Labeling of Glycosylated Anti-HER2 scFv Fusion Protein

1. AlexaFluor® 488C5-aminooxyacetamide,bis(triethylammonium) salt, 1 mg/mL in DMSO (Invitrogen). 2. 3 M Sodium acetate buffer, pH 5.0.

3. Methods 3.1. Degalactosylation of Monoclonal Antibodies

The degree of heterogeneity of the oligosaccharides in the native mAbs was initially determined by carrying out MALDI-TOF mass spectrometry (MS) analysis of the oligosaccharides released after PNGase F treatment of the mAbs (see Subheading  3.3 below). For example, mass spectrometry analysis of the oligosaccharides released after PNGase F treatment of the native form of Herceptin showed that it has only the G0 and G1 glycoforms (Fig. 1, left). Furthermore, none of the mAbs used in this work had sialylated structures as is observed with many other mAbs (6).

Fig. 1. Confirmation of the selective labeling of asialo-agalacto IgG molecules by MALDI-TOF MS analysis of PNGase F-treated samples. (a) Schematic of the native IgG molecule with desialylated N-glycan chains at Asn 297. MALDI-TOF MS analysis of the PNGase F-treated IgG sample shows a molecular mass of 1,485.5 m/z, the G0 glycoform of IgG, a molecular mass of 1,647.5 m/z corresponding to the G1 glycoform of IgG and 1,890 m/z for the G2 glycoform. (b) MALDITOF MS analysis of a sample of native IgG treated with b1,4-galactosidase (from S. pneumoniae). The PNGase F-treated degalactosylated IgG sample shows a single G0 glycoform corresponding to a molecular mass of 1,485.9 m/z. (c) The fully degalactosylated IgG contains four GlcNAc residues per IgG molecule available for the transfer of C2-keto-Gal from its UDP-derivative by Y289L-b4Gal-T1. After PNGase F treatment and MALDI-TOF MS analysis, the N-glycan of IgG shows a shift in the molecular mass to 1,890 m/z, indicating that the C2-keto-Gal has been transferred to both arms of the IgG molecule. The mannose residues in the N-glycan of the IgG are shown as darker circles, the GlcNAc residues as squares, the fucose residue as triangles, the galactose residue as lighter circles, and the C2-keto-galactose as ovals.

Bioconjugation Using Mutant Glycosyltransferases

287

To selectively remodel the oligosaccharide of the monoclonal antibodies at Asn 297 in the Fc domain and thereby obtain a homogeneous population of glycoforms, the mAbs were degalactosylated using recombinant b1,4-galactosidase from S. pneumoniae (Calbiochem) for 24 h (see Note 2). 1. Dissolve 1 mg of mAb in 500 mL of 50 mM sodium phosphate buffer, pH 6.0. 2. Transfer the protein solution to a Microcon YM-50 spin column and concentrate it to 100  mL by centrifuging at 10,000 × g. Repeat this step twice. 3. Incubate the mAb protein (8  mg/mL) with 100  mU of recombinant S. pneumoniae b1,4-galactosidase for 24  h at 37°C. 4. Purify the degalactosylated monoclonal antibodies by Protein A affinity chromatography (see Subheading 3.2). 3.2. Protein A Affinity Chromatography of Degalactosylated mAbs

1. 500  mL of resuspended Protein A-Sepharose® media was washed with ten volumes of 1× PBS, pH 7.4. 2. Degalactosylated samples of the mABs (obtained from Subheading 3.1) were diluted 1:1 (v/v) with binding buffer (1× PBS, pH 7.4) and loaded onto the Protein A column. 3. The column was washed twice with five volumes of 1× PBS, pH 7.4. 4. The bound mAbs were eluted with 100  mM glycine–HCl, pH 2.7. 5. The eluted mAbs were neutralized with 1 M Tris–HCl buffer, pH 8.0. 6. The eluted mAbs were concentrated and washed with 1× PBS (pH 7.4) using a Microcon YM-50 centrifugal filter device. 7. The amount of protein eluted from the affinity column was determined using the Bio-Rad Protein Assay kit. The final protein concentration of the eluted solution was adjusted to 1 mg/mL. 8. The purity of the eluted mAbs was assessed further by SDSPAGE analysis.

3.3. Mass Spectrometry Analysis of Oligosaccharides Released After PNGase F Treatment of Degalactosylated mAbs

The degree to which the mAbs were degalactosylated (see Subheading 3.1) was confirmed by MALDI-TOF MS analysis of the N-glycans released after treatment with PNGase F. 1. The following were added to a clean tube: 1 mL of 1 mg/mL of mAbs (native or degalactosylated), 1 mL of 10× G7 buffer, 1 mL of PNGase F (2,500 U), and 7 mL of water. 2. The reaction mixture was incubated for 16 h at 37°C.

288

Ramakrishnan et al.

3. The released N-glycans in the sample were purified using a micro-spin active charcoal column (see Note 3), and eluted with 30% (v/v) acetonitrile in water. 4. The released N-glycans were analyzed by mass spectrometry. 3.4. Transfer of C2-Keto-Galactose to Degalactosylated mAbs or Asialofetuin Using Engineered Glycosyltransferases

The following procedure was used to transfer C2-keto-Gal to free GlcNAc residues in degalactosylated monoclonal antibodies using the Y289L-b4Gal-T1 mutant enzyme, and also to transfer C2-keto-Gal to free LacNAc residues in asialofetuin using the 280 SGG282-a3Gal-T mutant enzyme. 1. To transfer C2-keto-Gal to free GlcNAc residues in the N-glycans of degalactosylated mAbs, the following were added to a clean tube: 5  mL of 2  mg/mL degalactosylated mAbs, 5 mL of 10 mM UDP-C2-keto-galactose, 6.3 mL of a 2.5  mg/mL stock solution of Y289L-b4Gal-T1 mutant enzyme, 1.25 mL of 500 mM Tris–HCl, pH 8.0, 1.25 mL of 100 mM MnCl2, and 11.2 mL of water. 2. To transfer C2-keto-Gal to free LacNAc residues in the N-glycans of asialofetuin, the following were added to a clean tube: 5 mL of 20 mg/mL asialofetuin, 5 mL of 10 mM UDPC2-keto-galactose, 6.3 mL of a 2.5 mg/mL stock solution of 280 SGG282-a3Gal-T mutant enzyme, 1.25  mL of 500  mM Tris–HCl, pH 8.0, 1.25 mL of 100 mM MnCl2, and 11.2 mL of water. 3. The glycosyltransferase reactions were incubated at 30°C for 3 h or overnight. 4. The ketone-labeled protein products obtained from the above glycosylation reactions were directly used for subsequent conjugation reactions (see Subheadings 3.5 and 3.7) without further purification. We followed the transfer of C2-keto-Gal to mAbs by MALDITOF MS analysis of the PNGase F-released oligosaccharides. Nearly 100% conversion of 80 pmol of the G0 form of IgG (12 mg) to the G2 form with C2-keto-Gal (Fig.  1, right) was achieved using a 2-mM concentration of the UDP-C2-keto-Gal sugar donor substrate and ~360  pmol of Y289L-b4Gal-T1 mutant enzyme (12 mg) in a 25-mL incubation mixture. Similarly, we also studied the transfer of C2-keto-Gal to the LacNAc moieties of the glycoprotein asialofetuin (which is known to have N-linked glycans with terminal LacNAc residues) by the 280SGG282-a3Gal-T mutant enzyme. The MALDI-TOF MS analysis of the N-glycans released after the PNGase F treatment of asialofetuin show biantennary (1,664 m/z) and triantennary (2,029 m/z) structures as reported previously (Fig. 2a) (3). Figure 2b shows the N-glycans from asialofetuin after the transfer of C2-keto-Gal by the mutant

Bioconjugation Using Mutant Glycosyltransferases

289

Fig.  2. Mass spectrometry (MS) analysis of the N-glycans released from asialofetuin after PNGase F treatment. The released oligosaccharides were passed through a spin column with active charcoal and analyzed by MALDI-TOF MS. Only the peaks of interest are annotated, showing their molecular mass and possible structures. All carbohydrate structures are shown in symbol form, as described in the legend to Fig. 1. (a) Glycans released from asialofetuin before C2-keto-Gal transfer. Ions at 1,664 m/z (biantennary galactosylated N-glycans) and 2,029 m/z (triantennary galactosylated N-glycans) are assigned to N-glycan structures reported previously for asialofetuin (3). (b) Glycan structures released from asialofetuin after transfer of C2-keto-Gal from UDP-C2-keto-Gal with the mutant enzyme 280SGG282-a3Gal-T. Ions at 1,866 and 2,068 m/z indicate the mass of biantennary N-glycans with transferred C2-keto-Gal on one and two arms, respectively. Ions at 2,231, 2,433, and 2,635 m/z indicate the mass of triantennary N-glycans with one, two, and three transferred C2-keto-Gal residues to the three arms, respectively.

enzyme 280SGG282-a3Gal-T to one (1,886 m/z) or both (2,086 m/z) arms of the biantennary glycan, and to one (2,231 m/z), two (2,433 m/z), or all three (2,636 m/z) arms of the triantennary structure (Fig. 2b). After the transfer of C2-ketogalactose to LacNAc residues on the N-glycans of asialofetuin by the mutant 280SGG282-a3Gal-T, the ketone moiety at the C-2 position of galactose was coupled with AOB (see Subheading 3.5 and Fig. 3) and developed by chemiluminescence methods, as described below in Subheading 3.6. 3.5. Biotinylation of C2-Keto-GalactoseModified mAbs or Asialofetuin

1. To biotinylate C2-keto-galactosylated mAbs or asialofetuin, the following were added to a clean tube: 10  mL of ketonelabeled proteins (obtained from Subheading  3.4), 1.5  mL of 3 M sodium acetate buffer (pH 3.9), 3 mL of 30 mM N-amin ooxymethylcarbonylhydrazino-D-biotin, and 15.5  mL of water. 2. The biotinylation reactions were incubated with gentle shaking for 12–16 h at 25°C. The successful coupling of the biotinylated aminooxy ligand to the ketone group at the C2 position of the modified galactose sugar was detected by using a sensitive chemiluminescence assay (see Subheading 3.6).

290

Ramakrishnan et al.

Fig. 3. Chemoenzymatic detection of C2-keto-Gal transferred to asialofetuin. The transfer of C2-keto-galactose to LacNAc residues on the N-glycan chains of asialofetuin by the a3Gal-T mutant enzyme, 280SGG282-a3Gal-T, was monitored by linking with N ¢-aminooxymethylcarbonylhydrazino-D-biotin (AOB), followed by Western blotting and chemiluminescence detection. Chemiluminescence was detected only in the sample that contained UDP-C2-keto-Gal, mutant enzyme and 25  ng of asialofetuin (Lane 1). After the transfer of C2-keto-Gal, the asialofetuin samples were treated with PNGase F (+), which removes the N-glycan chains from the protein. In contrast to the untreated samples (−), the PNGase F-treated samples (+) exhibited no chemiluminescence (Lane 2 ), indicating that the transfer of C2-keto-Gal is selective for the glycan portion of asialofetuin.

3.6. SDS-PAGE/ Western Blotting Analysis of BiotinLabeled mAbs or Asialofetuin

Biotin-labeled protein samples (obtained from Subheading 3.5) were analyzed by SDS-PAGE, followed by Western blotting and chemiluminescent detection using the streptavidin–HRP ­technique. The procedure below describes the analysis of proteins using a protein gel electrophoresis system from Invitrogen, but can be easily modified for use with other systems. 1. 20  mL of the biotinylated protein sample was mixed with 20  mL of 2× Tris–glycine–SDS sample buffer containing b-mercaptoethanol and boiled for 10 min. 2. Various amounts of samples, starting from 10  ng of total ­protein, were loaded onto a 14% SDS-PAGE Tris–glycine gel, and the proteins were resolved by operating the gel electrophoresis system according to the manufacturer’s instructions. 3. The resolved proteins were transferred from the polyacrylamide gel to a nitrocellulose membrane (0.45-mm pore size) for Western blot analysis.

Bioconjugation Using Mutant Glycosyltransferases

291

4. The nitrocellulose membrane (blot) was blocked for 1  h at room temperature with shaking in blocking solution containing 5% (w/v) dry milk in 0.02% (v/v) Tween-20. 5. The blot was rinsed three times (5 min for each wash) with water. 6. The blot was probed with streptavidin-conjugated horseradish peroxidase (HRP) diluted 1:4,000 (v/v) in 20 mL of 1× PBS (pH 7.4), 0.02% (v/v) Tween-20 containing 3% (w/v) BSA for 1 h at 25°C. 7. The blot were washed four times (10 min for each wash) in 5% (w/v) dry milk containing 0.02% (v/v) Tween-20. 8. The chemiluminescence signal generated by the streptavidin– HRP conjugate was detected after a 2-min incubation of the membrane blot with a 1:1 (v/v) mixture of ECL™ Detection Reagents 1 and 2, followed by exposure to Kodak® Biomax™ XAR film. The lack of a chemiluminescence band in the Western blot obtained after PNGase F treatment of Asialofetuin (Fig. 3) and the mAbs (Fig. 4a) supports the conclusion that the biotinylated

Fig. 4. Detection of biotin tags and fluorescent probe molecules conjugated to the C2-keto-Gal residue transferred to free GlcNAc residues on the N-glycan chains of monoclonal antibodies (mAbs). (a) Chemiluminescent detection of C2-ketoGal-linked IgG conjugated to aminooxy-biotin. Biotin-labeled protein samples treated with and without PNGase F were resolved by SDS-PAGE, transferred to nitrocellulose membranes for Western blot analysis, and then probed with streptavidin–HRP conjugate. The sample not treated with PNGase F (−) shows chemiluminescence from the heavy chain component of the IgG. No chemiluminescence was detected in the PNGase F-treated sample (+), showing that the transfer of the modified sugar C2-keto-Gal is selective for the GlcNAc residues on the N-glycan chains. (b) C2-keto-Gal-modified mAb was conjugated with Alexa Fluor® 488 C5-aminooxyacetamide and analyzed by SDS-PAGE, followed by fluorescence imaging. (c) The same gel was subsequently stained with Coomassie blue dye. Lane M: Kaleidoscope protein molecular weight standards; Lane 1: Y289L-b4Gal-T1; Lane 2: C2-keto-Gal-modified Herceptin mAb conjugated to Alexa Fluor® 488 C5-aminooxy acetamide; Lane 3: C2-keto-Gal-modified Herceptin mAb conjugated to Alexa Fluor® 488 C5aminooxyacetamide after digestion with PNGase F; and Lane 4: PNGase F alone.

292

Ramakrishnan et al.

aminooxy ligand is linked only to the C2-keto-Gal-modified N-glycan chain(s) of Asialofetuin or of the IgG heavy chain, respectively. 3.7. Fluorescent Labeling of C2-KetoGalactose-Modified mAbs

1. To fluorescently label C2-keto-galactosylated mAbs, the following were added to a clean tube: 14.7 mL (10 mg) of C2-ketoGal-labeled Herceptin (or other mAb), 10 mL of Alexa Fluor® 488 C5-aminooxyacetamide, bis(triethylammonium) salt (1  mg/mL in DMSO), 1.5  mL of sodium acetate (pH 4.9), and 0.8 mL of water (see Note 4). 2. The reaction was incubated for 6 h overnight in the dark at room temperature. 3. Unreacted fluorescent dye molecules were removed by purification on a Micro Bio-Spin 30 chromatography column prepacked with Bio-Gel P-30, as per the manufacturer’s instructions. 4. The purified samples were separated by SDS-PAGE, and the fluorescence was detected using a multiview FMBIO II fluorescence imaging scanner. The conjugation of C2-keto-Gal-modified Herceptin with Alexa Fluor® 488 C5-aminooxyacetamide was analyzed by fluorescence imaging of the SDS-PAGE gel (Fig. 4b), and subsequent staining of the gel with Coomassie blue (Fig.  4c). The fluorescence imaging analysis indicates that there is a selective labeling of the heavy chain of the mAb, and that the resultant fluorescence signal intensity significantly decreases following treatment of the fluorescent-tagged antibody with PNGase F, which removes the N-linked sugars in the Fc domain of the IgG. As expected, no fluorescence was detected from other protein bands that were visible after Coomassie blue staining of the SDS-PAGE gel.

3.8. Glycosylation of Anti-HER2 scFv Fusion Protein

The anti-HER2 scFv fusion protein incorporates a Tobacco Etch Virus (TEV) cysteine protease cleavage site, and an O-glycosylation peptide that is followed by a 6× His-tag at its C-terminal end. The O-glycosylation peptides used in this work contained one or multiple glycosylation sites, and were designed based on the preferred acceptor peptide sequence for the h-ppaGalNAc-T2 enzyme (described in Subheading 1.2). In the procedure described below, h-ppaGalNAc-T2 is used to transfer C2-keto-Gal from the UDPC2-keto-Gal donor substrate to specific O-glycosylation sites in the anti-HER2 scFv fusion protein. 1. To transfer C2-keto-Gal to specific O-glycosylation sites in the anti-HER2 scFv fusion protein, the following were added to a clean tube: 40 mL of 1 mg/mL anti-HER2 scFv fusion protein, 20 mL of 1 mg/mL of h-ppaGalNAc-T2 enzyme, 10 mL of 250 mM Tris–HCl (pH 8.0), 10 mL of 100 mM MnCl2, 5 mL of 10 mM UDP-C2-keto-Gal, and 15 mL of water.

Bioconjugation Using Mutant Glycosyltransferases

293

2. The reaction mixture (100  mL) was incubated overnight at room temperature ~20°C. A small amount of the glycoprotein product obtained from the above glycosylation reaction was subjected to TEV digestion, and the released C-terminal fusion glycopeptides were analyzed by MALDI-TOF MS. The results showed that scFv fusion proteins containing a single O-glycosylation site were fully glycosylated with a single C2-keto-Gal sugar (10). 3.9. Purification of Glycosylated Anti-HER2 scFv Fusion Protein

To remove the reaction buffer components from the glycosylation reactions (Subheading  3.8), the glycosylated anti-HER2 scFv fusion protein is first precipitated with ammonium sulfate and then desalted as follows: 1. Add 400  mL of water to 100  mL of the reaction mixture (obtained from Subheading  3.8) containing glycosylated anti-HER2 scFv fusion protein. 2. Add 0.5 g of ammonium sulfate and gently vortex the mixture until the ammonium sulfate crystals are fully dissolved. 3. Spin at 10,000 × g to collect the protein precipitates and discard the clear supernatant. 4. Dissolve the recovered protein precipitate in 500 mL of water, and then concentrate the sample to a total volume of 20 mL using a Microcon YM-10 spin column at 10,000 × g. 5. Dilute and concentrate the protein solution twice by adding 480 mL of water and then concentrating the sample to 20 mL at 10,000 × g. 6. Reconstitute the final, salt-free glycosylated anti-HER2 scFv fusion protein in 40 mL of water to give a final protein concentration of 1 mg/mL.

3.10. Fluorescent Labeling of Glycosylated AntiHER2 scFv Fusion Protein

Anti-HER2 scFv fusion proteins modified with C2-keto-Gal sugars (Subheading 3.9) were conjugated with a fluorescence probe, Alexa Fluor® 488, having an orthogonal reactive chemical group, viz., the aminooxy group. 1. To fluorescently label the glycosylated anti-HER2 scFv fusion protein, the following were added to a tube: 40 mL of 1  mg/mL desalted C2-keto-Gal-modified anti-HER2 ScFv fusion protein, 3  mL of 3  M sodium acetate buffer (pH 5.0), 8 mL of Alexa Fluor® 488 C5-aminooxyacetamide, bis(triethylammonium) salt (1 mg/mL in DMSO), and 9 mL of water. 2. The reaction mixture was incubated in the dark overnight at room temperature.

294

Ramakrishnan et al.

3. The conjugated protein was precipitated with ammonium sulfate by first diluting the reaction mixture to 500 mL with water, and then using the procedure described in Subheading 3.9 above. 4. The purified samples were separated by SDS-PAGE, and the fluorescence was detected using a multiview FMBIO II fluorescence imaging scanner. As shown in Fig. 5, the conjugated protein should be detectable on an SDS-PAGE gel by its fluorescence emission, even at nanogram quantities.

Fig. 5. Fluorescent labeling of a single-chain antibody against the human HER2 receptor (anti-HER2 scFv) fused to a C-terminal peptide-tag containing a single or multiple O-glycosylation sites. In the presence of Mn2+, the human polypeptide-a-N-acetylgalactosaminyltransferase II (h-ppaGalNAc-T2) enzyme can transfer a C2-keto-Gal sugar from its respective UDP derivative to the side chain hydroxyl group of Thr residues present in the O-glycosylation tag. Alexa Fluor® 488 can be conjugated to the C2-keto-Gal label in the C-terminal peptide-tag of the anti HER2 scFv fusion protein, and the fluorescence emission visualized from a (nonreduced) SDS-PAGE gel. The lanes in the gel show the fluorescence emission obtained from loading 10, 20, and 40 ng of Alexa Fluor® 488-labeled scFv fusion protein. Even at the lowest amount of fusion protein loaded (i.e., 10 ng), the protein band remains sharp and is easily detectable.

Bioconjugation Using Mutant Glycosyltransferases

295

4. Notes 1. Unless stated otherwise, all buffers and solutions are prepared using high-purity water (referred to as “water” throughout the text) that has a resistivity of 18.2 MW⋅cm at 25°C. 2. The mAbs are degalactosylated using recombinant b1, 4-galactosidase from S. pneumoniae (Calbiochem). We have observed that a 24-h incubation of 8 mg/mL of mAbs with 100 mU of b1,4-galactosidase from S. pneumoniae completely converts the mAbs to the G0 glycoform; however, b1, 4-galactosidase obtained from other microorganism sources (e.g., E. coli) (or other commercial suppliers) did not achieve the same result. 3. The use of charcoal columns is essential to purify the modified glycans after PNGase F treatment of the glycoproteins and prior to MALDI-TOF MS analysis. 4. The conjugation reaction between the keto-labeled samples and the Alexa Fluor® 488 C5-aminooxyacetamide, bis(triethylammonium) salt is very sensitive to pH. The conjugation only takes place efficiently at pH ~4.8–5.0. References 1. Ramakrishnan, B. and Qasba, P. K. (2002) Structure-based design of beta-1,4-galactosyltransferase-I (beta 4Gal-T1) with equally efficient N-Acetylgalactosaminyltransferase activity: point mutation broadens beta 4Gal-T1 donor specificity. J. Biol. Chem. 277, 20833–20839. 2. Khidekel, N., Arndt, S., Lamarre-Vincent, N., Lippert, A., Poulin-Kerstien, K. G., Ramakrishnan, B., Qasba, P. K., and HsiehWilson, L. C. (2003) A chemoenzymatic approach toward the rapid and sensitive detection of O-GlcNAc posttranslational modifications. J. Am. Chem. Soc. 125, 16162–16163. 3. Pasek, M., Ramakrishnan, B., Boeggeman, E., Manzoni, M., Waybright, T. J., and Qasba, P. K. (2009) Bioconjugation and detection of lactosamine moiety using alpha1,3-galactosyltransferase mutants that transfer C2-modified galactose with a chemical handle. Bioconjug. Chem. 20, 608–618. 4. Carter, P. J. and Senter, P. D. (2008) Antibodydrug conjugates for cancer therapy. Cancer J. 14, 154–169. 5. Stimmel, J. B., Merrill, B. M., Kuyper, L. F., Moxham, C. P. and Hutchins, J. T. (2000) Site-specific conjugation of serine → cysteine

variant monoclonal antibodies. J. Biol. Chem. 275, 30445–30450. 6. Dhawan, S. (2002) Design and construction of novel molecular conjugates for signal amplification (I): conjugation of multiple horseradish peroxidase molecules to immunoglobulin via primary amines on lysine peptide chains. Peptides 23, 2091–2098. 7. Boeggeman, E., Ramakrishnan, B., Kilgore, C., Khidekel, N., Hsieh-Wilson, L. C., Simpson, J. T., and Qasba, P. K. (2007) Direct identification of nonreducing GlcNAc residues on N-glycans of glycoproteins using a novel chemoenzymatic method. Bioconjug. Chem. 18, 806–814. 8. Ramakrishnan, B., Boeggeman, E. and Qasba, P. K. (2007) Novel method for in  vitro O-glycosylation of proteins: application for bioconjugation. Bioconjug. Chem. 18, 1912–1918. 9. Boeggeman, E., Ramakrishnan, B., Pasek, M., Manzoni, M., Puri, A., Loomis, K. H., Waybright, T. J. and Qasba, P. K. (2009) Sitespecific conjugation of fluoroprobes to the remodeled Fc N-glycans of monoclonal antibodies using mutant glycosyltransferases: application for cell surface antigen detection. Bioconjug. Chem. 20, 1228–1236.

296

Ramakrishnan et al.

10. Ramakrishnan, B., Boeggeman, E., Manzoni, M., Zhu, Z., Loomis, K., Puri, A., Dimitrov, D. S. and Qasba P. K. (2009) Multiple site-specific in  vitro labeling of single-chain antibody. Bioconjug. Chem. 20, 1383–1389. 11. O’Shanessy, D. J., Dobersen, M. J. and Quarles, R. H. (1984) A novel procedure for labeling immunoglobulins by conjugation to oligosaccharide moieties. Immunol. Lett. 8, 273–277. 12. Holland, M., Yagi, H., Takahashi, N., Kato, K., Savage, C. O., Goodall, D. M. and Jefferis, R. (2006) Differential glycosylation of polyclonal IgG, IgG-Fc and IgG-Fab isolated from

the sera of patients with ANCA-associated systemic vasculitis. Biochim. Biophys. Acta. 1760, 669–677. 13. Dube, D. H., Prescher, J. A., Quang, C. N. and Bertozzi, C. R. (2006) Probing mucintype O-linked glycosylation in living animals. Proc. Natl. Acad. Sci. U S A. 103, 4819–4824. 14. Boeggeman, E., Ramakrishnan, B. and Qasba, P. K. (2003) The N-terminal stem region of bovine and human b1,4-galactosyltransferase I increases the in  vitro folding efficiency of their catalytic domain from inclusion bodies. Prot. Expres. Purif. 30, 219–220.

Chapter 18 Lipid-Core-Peptide System for Self-Adjuvanting Synthetic Vaccine Delivery Mariusz Skwarczynski and Istvan Toth Abstract Disadvantages of classical vaccines, such as the risk of an autoimmune reaction, might be overcome by using a subunit vaccine containing the minimal microbial components necessary to stimulate appropriate immune responses. However, vaccines based on minimal epitopes suffer from poor immunogenicity and require the use of an additional immunostimulant (adjuvant). Only a few adjuvants have been permitted for use with vaccines intended for human administration. We have developed several vaccine candidates based on a lipid-core-peptide (LCP) system. This system has self-adjuvanting properties, and it can be used for the delivery of a variety of epitopes to produce vaccine candidates against a targeted disease. The LCP system is easily assembled by simple stepwise Boc solid-phase peptide synthesis. Key words: Lipoamino acids, Synthetic vaccine, Lipid-core-peptide, Adjuvant, Boc solid-phase peptide synthesis, Multiple antigenic peptide

1. Introduction The lipid-core-peptide (LCP) delivery system is a promising approach for human vaccination that has been developed based on the principle of lipopeptide vaccines. The LCP system contains a nonmicrobial lipopeptide adjuvant based on lipoamino acids (LAAs) and a polylysine branching scaffold, which provides conjugation sites for the peptide epitopes (1, 2). The whole self-adjuvanting system can be easily assembled by solid-phase peptide synthesis (SPPS). We have also reported several modifications to this platform, including the application of native chemical ligation (3–6), branching carbohydrate scaffolds (3, 4, 7), changing the level of polylysine branching (8, 9), and the length and arrangement of LAAs (10). Nevertheless, the basic system

Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_18, © Springer Science+Business Media, LLC 2011

297

298

Skwarczynski and Toth

peptide epitope

peptide epitope

N H

N H

O

H N

N H

O

HN peptide epitope

N H

peptide epitope

N H

H N

O N H

O

H N

O

O

N H

NH 2 O

O

LCP core

Fig. 1. Typical self-adjuvanting LCP system possessing four peptide epitopes.

presented in Fig. 1 remains the first choice for the study of new vaccine candidates. This protocol demonstrates the synthesis of the LCP system for a vaccine candidate against Group A streptococcal (GAS) infection. The procedure includes three major parts:(1)thesynthesisofaprotectedLAA(2-(R/S)-aminodo­decanoic acid), (2) the synthesis of the LCP core, and (3) the final arrangement of the whole construct (LCP-J8) (11–13), including multiple copies of the GAS B-cell epitope, J8 (see Fig.  1; peptide epitope = QAEDKVKQSREAKKQVEKALKQLEDKVQ). We synthesized the construct by manual Boc SPPS; however, microwaveassisted or automatic Boc SPPS may also be employed (14).

2. Materials 2.1. Synthesis of 2-(R/S)-[(tertButoxycarbonyl) amino]-dodecanoic acid

1. Diethyl acetamidomalonate. 2. Bromodecane. 3. Sodium. 4. Di-tert-butyl dicarbonate (Boc2O). 5. Solvents: ethanol, dimethylformamide (DMF), acetone, dioxane, ethyl acetate, acetonitrile, and deuterated chloroform (CDCl3). 6. Hydrochloric acid (HCl), 32–35% (w/w). 7. Aqueous ammonium hydroxide solution, 28% (w/w). 8. Sodium hydroxide.

Lipid-Core-Peptide System for Self-Adjuvanting Synthetic Vaccine Delivery

299

9. Sodium chloride saturated solution (brine). 10. Activated charcoal. 11. Anhydrous magnesium sulfate. 2.2. Synthesis of LCP

1. p-Methylbenzhydrylamine (p-MBHA) resin⋅HCl (substitution about 0.45 meq/g). 2. t-Butyloxycarbonyl (t-Boc) protected amino acids. 3. Dimethylformamide (DMF). 4. Dichloromethane (DCM). 5. N,N-Diisopropylethylamine (DIPEA). 6. Trifluoroacetic acid (TFA). 7. O-Benzotriazole-N,N,N¢,N¢-tetramethyl-ur oniumhexafluoro-phosphate (HBTU) solution: 0.5  M HBTU in DMF. Store at 0°C (see Note 1). 8. Capping solution: Add 0.5  mL of acetic anhydride and 0.5 mL of DIPEA to 10 mL of DMF (10 mL).

2.3. Arrangement of the Peptide Epitopes on LCP

1. All materials as described in Subheading 2.2 above. 2. Ninhydrin (Kaiser) test solutions. Reagent A: Dissolve 76 g of phenol in 24  g of ethanol; Reagent B: Dissolve 132  mg potassium cyanide (KCN) in a small quantity of water (~2 mL) and add 98 mL pyridine; and Reagent C: Dissolve 5 g of ninhydrin in 100 mL of ethanol using an ultrasonic bath. Store all reagents under dark at 2–8°C. 3. Solution A: Acetonitrile/water/TFA, 50/50/0.1 (v/v/v). 4. Methanol. 5. p-Cresol.

2.4. Equipment

1. Nuclear magnetic resonance (NMR) spectrometer (Bruker Avance™ 300 MHz). 2. Apparatus for HF cleavage. 3. Lyophilizer. 4. HPLC system equipped with preparative and analytical C4 columns. 5. Mass spectrometer (Perkin-Elmer-Sciex API 3000). 6. Peptide synthesis vessel. 7. Magnetic stirrer with hot plate, magnetic stir bar. 8. Desiccator. 9. Laboratory glassware.

300

Skwarczynski and Toth HCl H 2N

Br

+ O

1. EtONa O

O

O OH Boc2O

H N

O

O OH

O

2. HCl O

HN O

Fig. 2. Synthesis of 2-(R/S)-[(tert-butoxycarbonyl) amino]-dodecanoic acid (C12).

3. Methods The LCP system uses versatile LAAs and their flexible combinations. LAAs are a-amino acids that contain alkyl side chains of a desired length. One of the most popular LAAs used in the LCP system is 2-(R/S)-aminododecanoic acid (C12) with an N-tertbutoxycarbonyl (N-Boc) protective group. The synthesis of other LAAs can be achieved in exactly same manner as C12 (see Fig. 2) (15). Following N-Boc deprotection, the LAAs, a Gly spacer, and Boc-Lys(Boc) are sequentially added to the resin using a modified version of the in situ neutralization protocol for Boc SPPS (16, 17). Finally, the epitopes are incorporated into the LCP by stepwise SPPS (Fig. 1). 3.1. Synthesis of 2-(R/S)-[(tertButoxycarbonyl) amino]-dodecanoic acid (Boc-C12)

1. Preparation of sodium ethoxide: To a 500-mL round-bottom 0.71  mol) and then add sodium (1.15  g, 0.05  mol) (see Note 2). 2. To the above solution, add diethyl acetamidomalonate (10.9 g, 0.05 mol) and bromodecane (14.5 mL, 0.07 mol). Attach a reflux condenser to the flask and gently reflux the mixture for 1 day with stirring. 3. Cool the reaction mixture to room temperature. 4. Pour the reaction mixture into ice water and filter the precipitate using a Buchner funnel under vacuum. Wash the solid with cold water (5 × 50 mL) (see Note 3) and dry overnight in a desiccator under reduced pressure. 5. Transfer the solid to a 500-mL round-bottom flask equipped with magnetic stir bar. 6. Add DMF (7  mL) and hydrochloric acid (32–35%, w/w) (60 mL) and reflux as in step 2 for 3 days. 7. Cool the brownish-black reaction mixture to room temperature.

Lipid-Core-Peptide System for Self-Adjuvanting Synthetic Vaccine Delivery

301

8. Add slowly with stirring aqueous ammonium hydroxide solution (28%, w/w) (60 mL) and stir for 5 min (see Note 4). 9. Pour the reaction mixture into ice water and filter the precipitate using a Buchner funnel under vacuum. Wash the solid with cold water (5 × 50  mL) and cold acetone (1 × 50  mL) and dry overnight in a desiccator under reduced pressure. The final 2-(R/S)-aminododecanoic acid hydrochloride product is a white or light brown solid (9.09 g, 0.036 mol, yield = 72%). The product (2-(R/S)-aminododecanoic acid hydrochloric acid salt) was analyzed by mass spectrometry. Electrospray ionization mass spectrometry (ESI-MS): m/z 216 [M + H]+. 10. Transfer the 2-(R/S)-aminododecanoic acid hydrochloric acid salt compound (9.09 g, 0.036 mol) to a 500-mL roundbottom flask equipped with magnetic stir bar. 11. Dissolve sodium hydroxide (3.38  g, 0.084  mol) in water (100 mL), and allow the solution to cool down before adding it to the flask. Stir the mixture until the LAA dissolves completely. 12. Add di-tert-butyl dicarbonate (Boc2O) (9.22 g 0.042 mol) in dioxane (100 mL) to the mixture and stir at room temperature for 3–5 h (see Note 5). 13. Evaporate the dioxane under reduced pressure and acidify the remaining aqueous solution to pH 3 with 5% cold hydrochloric acid (see Note 6). 14. Extract the solution with ethyl acetate (2 × 300 mL) using a separation funnel, and then combine the organic fractions and wash with brine (1 × 50 mL) (see Note 7). 15. Mix the ethyl acetate solution with activated charcoal, filter the charcoal off, add anhydrous magnesium sulfate to dry the solution, filter the magnesium sulfate off, and then evaporate the ethyl acetate under reduced pressure. 16. To the resultant oil, add a small amount of acetonitrile (5–20 mL), cool the flask down to −10 to 0°C (in a sodium salt/ice water bath), and scratch the resultant oily solid with a metal spatula to obtain a white solid. 17. Wash the solid with cold acetonitrile (1 × 5 mL) and filter the product using a Buchner funnel under vacuum. The final product (2-(R/S)-[(tert-butoxycarbonyl) amino]-dodecanoic acid, Boc-C12) is a white solid (5.17 g, 0.016 mol, yield = 45%) (see Note 8). The product was analyzed by ESI-MS and NMR spectroscopy. ESI-MS: m/z 316 [M + H]+. 1H NMR (300 MHz, CDCl3): d 0.86 (t, 3H, J = 7.0  Hz, CH3), 1.24 (bs, 16H, 8 × CH2), 1.43 (s, 9H, CMe3), 1.56–1.85 (m, 2H, b-CH2), 4.30–4.20 (m, 1H, a-CH), 4.93 (d, 1H, J 7.1 Hz, NH).

302

Skwarczynski and Toth

3.2. Synthesis of LCP

1. Resin swelling: Transfer the p-MBHA resin⋅HCl (for 0.45 meq/g substituted resin: 0.22 g, 0.1 mmol) to the peptide synthesis vessel (separation funnel with a fritted disc). Add dry DMF (5  mL) and DIPEA (0.11  mL, 0.62  mmol) and gently shake for 1–2 h at room temperature (see Note 9). 2. Remove the solvent by filtration under vacuum, wash under flow with DMF (2 × 5 mL) by adding the DMF continuously and draining under vacuum (see Note 10). 3. Preactivation of amino acid: To a scintillation vial add BocGly-OH (73.6 mg, 0.42 mmol, 4.2 eq), then add a solution of HBTU (0.8 mL, 4.0 mmol, 4.0 eq) and DIPEA [0.11 mL, 0.62  mmol, 6.2  eq (6.2  eq = 1  eq (resin) + 4.2  eq (amino acid) + 1.0 eq (excess))]. Wait until the reagents dissolve completely, or use an ultrasonicator (often required) to speed up the dissolution process. Preactivation should be performed 2–5 min before proceeding to the next step. 4. Add the preactivated amino acid to the resin and shake the mixture for 60 min. 5. Remove the reaction mixture by filtration under vacuum, wash under flow with DMF (2 × 5 mL) by adding DMF continuously and draining under vacuum. 6. Repeat steps 3 and 4 for double coupling. 7. Remove the reaction mixture by filtration under vacuum, wash under flow with DMF by adding DMF continuously and draining under vacuum (5 × 5 mL). 8. After the first amino acid is attached to the resin, capping is performed as a standard procedure. To the washed resin, add freshly prepared capping solution (2 mL). Shake the resin for 20 min. 9. Remove the reaction mixture by filtration under vacuum, and then repeat step 8. 10. Repeat step 7. 11. Boc deprotection: To the resin, add neat TFA (5 mL), and shake for 1 min. Remove the TFA by filtration under vacuum. Add the next portion of TFA (5 mL), and again shake for 1 min. 12. Remove the TFA by filtration under vacuum, and wash the resin under flow with DMF by adding DMF continuously and draining under vacuum (10 × 5 mL). 13. Repeat steps 3–7, 11, and 12 five more times with the following changes. Instead of Boc-Gly, use 0.42  mmol of the following Boc protected amino acids in sequence: Boc-C12, Boc-C12, Boc-Gly, Boc-C12, and Boc-Lys(Boc)-OH. BocC12 is less reactive and should be preactivated for approximately 5–10 min.

Lipid-Core-Peptide System for Self-Adjuvanting Synthetic Vaccine Delivery

303

14. Preactivation of the second Boc-Lys(Boc)-OH: To a scintillation vial add Boc-Lys(Boc)-OH (291 mg, 0.84 mmol), then add a solution of HBTU (1.6 mL, 8.0 mmol) and DIPEA (0.22 mL, 1.24 mmol). Proceed as described in step 3 (see Note 11). 15. Add the preactivated amino acid to the resin. Shake for 30 min. 16. Repeat steps 2, 14, and 15. Following this, then repeat steps 7, 11, and 12 (see Note 12). 3.3. Arrangement of the Peptide Epitopes on LCP with J8 Peptide as an Example

1. Preactivation of amino acid: To a scintillation vial, add BocGln(Xan)-OH (717 mg, 1.68 mmol), then add a solution of HBTU (3.2 mL, 1.6 mmol) and DIPEA (0.44 mL, 2.48 mmol) (see Note 13). Wait until the reagents dissolve completely, or use an ultrasonicator (often required) to speed up the dissolution process. Preactivation should be performed 2–5  min before proceeding to the next step. 2. Add the preactivated amino acid to the resin and shake the mixture for 30 min. 3. Perform a qualitative ninhydrin (Kaiser) test (see Note 14): Using a Pasteur pipette, remove ~0.1 mL of the suspension containing ~2–3  mg of resin. Wash the resin under flow with DMF, DCM, and methanol in a separate small filtration funnel. Allow the resin to dry for a few minutes in the funnel under vacuum. Next, transfer the resin to a 5-mL test tube. Prepare an empty test tube as a blank. To each tube, add two drops of Reagent A, four drops of Reagent B, and two drops of Reagent C. Incubate the mixture at 100°C for 5 min. If the beads and/or the solution become violet, blue, or slightly blue, double coupling should be performed (wash the resin, and repeat steps 1 and 2) as conversion of the free amine group of the terminal amino acid on the resin is below 99.5%. 4. Remove the reaction mixture by filtration, wash under flow with DMF (5 × 5  mL) by adding DMF continuously and draining under vacuum (see Note 15). 5. Boc deprotection: To the resin, add neat TFA (5  mL) and shake for 1 min. Remove the TFA by filtration under vacuum. Add the next portion of TFA (5  mL) and again shake for 1 min. 6. Remove the TFA by filtration under vacuum, and wash the resin under flow with DMF by adding DMF continuously and draining under vacuum (10 × 5 mL) (see Note 15). 7. Repeat steps 1–6 twenty-seven times with the following changes. Instead of Boc-Gln(Xan)-OH, use 1.68  mmol of the desired amino acids in the following order: Boc-Val-OH,

304

Skwarczynski and Toth

Boc-Lys(2Cl-Z)-OH, Boc-Asp(OcHx)-OH, Boc-Glu (OcHx)-OH, Boc-Leu-OH·H2O, Boc-Gln(Xan)-OH, BocLys(2Cl-Z)-OH, Boc-Leu-OH·H2O, Boc-Ala-OH, BocLys(2Cl-Z)-OH, Boc-Glu(OcHx)-OH, Boc-Val-OH, Boc-Gln(Xan)-OH, Boc-Lys(2Cl-Z)-OH, Boc-Lys(2Cl-Z)-OH, Boc-Ala-OH, Boc-Glu(OcHx)-OH, Boc-Arg(Tos)-OH, BocSer(Bzl)-OH, Boc-Gln(Xan)-OH, Boc-Lys(2Cl-Z)-OH, Boc-Val-OH, Boc-Lys(2Cl-Z)-OH, Boc-Asp(OcHx)-OH, BocGlu(OcHx)-OH, Boc-Ala-OH, and Boc-Gln(Xan)-OH (see Note 16). 8. Final acetylation: To the washed resin, add freshly prepared capping solution (2 mL). Shake the resin for 20 min. 9. Remove the reaction mixture by filtration under vacuum, and then repeat step 8. 10. Remove the reaction mixture by filtration under vacuum, and then wash the resin under flow with DMF by adding DMF continuously and draining under vacuum (5 × 5 mL). 11. Wash the resin under flow with DCM (5 × 5 mL) and methanol (1 × 5 mL). 12. Dry the resin overnight in a desiccator under reduced pressure. 13. Check the weight of the obtained resin. 14. For the cleavage of the construct from the resin, use a dedicated hydrofluoric acid (HF)-reaction apparatus. To the resin, add p-cresol scavenger (0.5 mL/g of resin) and HF (10 mL/g of resin). Proceed with HF-cleavage of the construct from the resin (see Note 17). 15. To the mixture obtained after the HF-cleavage operation (HF-reaction vessel), add 10 mL of cold diethyl ether (stored in a freezer), stir for 1 min, and remove the solvent by filtration under vacuum to remove the scavenger (see Note 18). 16. Repeat step 15 two more times. 17. Add Solution A (10 mL) to dissolve the obtained construct, and then filter off the resin. 18. Wash the HF-reaction vessel and resin with Solution A (2 × 10 mL). 19. Combine the fractions of Solution A and lyophilize. 20. Purify the obtained solid by preparative reversed-phase HPLC. The final product (LCP-J8) is a white powder. The mass spectrum of LCP-J8 has multiply charged ions: (M + 7H+)/7:2023.1 (calcd. 2023.6), (M + 8H+)/8:1771.3 (calcd. 1770.8), (M + 11H+)/11: 1288.9 (calcd. 1288.1), (M + 12H+)/12: 1179.7 (calcd. 1180.8), (M + 13H+)/ 13:1088.7 (calcd. 1090.1), (M + 14H+)/14: 1011.5 (calcd.

Lipid-Core-Peptide System for Self-Adjuvanting Synthetic Vaccine Delivery

305

1012.3), (M + 15H+)/15:943.8 (calcd. 944.9) (M + 16H+)/ 16: 885.4 (calcd. 885.9) (M + 17H+)/17: 833.8 (calcd. 833.8) (M + 18H+)/18: 788.3 (calcd. 787.6), (M + 19H+)/ 19:746.4 (calcd. 746.2). Compound can be stored at −20°C for several months.

4. Notes 1. This colorless solution should be used within a period of 1 or 2 weeks. If it turns to a yellow color, it should be discarded. Caution: HBTU may produce an allergic response after contact with skin. 2. Caution: Sodium reacts violently on contact with water and may cause fire or explosion. Sodium also reacts with ethanol at room temperature to liberate hydrogen. Therefore, the reaction should be carried out under a well-ventilated fume hood. To perform scaled-up reactions, the reaction flask should be cooled by ice water, and the sodium should be added in small portions. Prior to use, sodium should be washed with hexane. This preparation procedure can be omitted when commercially available sodium ethoxide is employed. 3. Keeping a low temperature during filtration is important, as the precipitate tends to change to an oily state at room temperature. 4. Partial neutralization of the mixture with ammonia helps in the precipitation of the product in step 9. 5. The pH of the mixture should be monitored so that it remains at pH = 9.5–10.5. If the pH is less basic after 1 h of reaction, pellets or a concentrated stock solution of sodium hydroxide should be added to adjust the pH. The progress of the reaction may be monitored by TLC (CHCl3/MeOH/AcOH, 90/8/2 v/v/v; product Rf » 0.5, starting material Rf » 0–0.1). The reaction generally does not proceed to full completion (at the end of the reaction, a trace amount of the starting material is still visible on TLC). 6. Due to the risk of having the Boc become partially deprotected under acidic conditions (5% HCl), the pH of the solution should not be lower than 3. 7. During extraction, mixture of the aqueous and organic phases tends to create an emulsion. To minimize this, carry out the extraction at a temperature close to 0°C. 8. Pure 2-(R/S)-[(tert-butoxycarbonyl) amino]-dodecanoic acid is a white odorless solid. Contamination from nonreacted Boc2O has to be avoided as it leads to substantial

306

Skwarczynski and Toth

byproduct formation during SPPS. Thus, a minimal amount or no excess of Boc2O should be employed for the synthesis reaction. The presence of Boc2O in the sample can be easily recognized by its characteristic smell or by 1H NMR spectra analysis. If Boc2O is detected, the sample should be recrystallized from acetonitrile. 9. Several different types of peptide reaction vessels and shakers are commercially available. Low-substituted resin should be used for the synthesis reaction (<0.5  meq/g). The solvent level should be high enough to ensure that the resin is coated and that there is good mobility during swelling and further reactions. Whenever necessary, more solvent should be added to maintain the appropriate reaction conditions. 10. Do not over-dry the resin during filtration. The resin should only be kept without solvent for the minimum amount of time necessary. 11. The required equivalents of reagents are doubled due to the first branching on the lysine. 12. If synthesis cannot be continued on the same day, the resin can be safely stored in DMF at 4°C for 1 week. However, the synthesis should be stopped before a Boc-deprotection step to avoid possible side reactions with the free amine group of the terminal amino acid. 13. The required equivalents of reagents are quadrupled due to the second branching on the lysine. 14. Alternatively, perform a quantitative ninhydrin test (18). The ninhydrin test can also be skipped at this step, and double coupling can be performed instead throughout the whole SPPS procedure. 15. Boc deprotection after coupling of Gln: The remaining DMF during treatment with TFA causes a rise in the reaction temperature of 70–80°C. Under these conditions, cyclization of the terminal glutamine moiety is a potentially serious side reaction issue that may terminate the growth of the peptide chain on the resin. To avoid this problem, before the first and after the second treatment with TFA, the resin should be washed with DCM (3 × 5 mL). 16. If an amino acid with a side chain protective group stable to HF cleavage is used, additional cleavage needs to be performed after step 19. 17. Caution: HF is highly toxic. Follow precisely the instructions attached to your HF apparatus. If your sequence contains cysteine or methionine, use a mixture of p-cresol and p-thiocresol (0.5 and 0.5  mL/g of resin, respectively) as a scavenger.

Lipid-Core-Peptide System for Self-Adjuvanting Synthetic Vaccine Delivery

307

18. For the filtration step, use polyethylene frits inserted into single-use plastic syringes (25 mL). Avoid the use of glass or porcelain funnels due to the presence of residual HF (HF can react with glass).

Acknowledgment This work was supported by the National Health and Medicinal Research Council (Australia). References 1. Zhong, W., Skwarczynski, M., and Toth, I. (2009) Lipid Core Peptide System for Gene, Drug, and Vaccine Delivery, Aust. J. Chem. 62, 956–967. 2. Moyle, P. M., and Toth, I. (2008) Selfadjuvanting lipopeptide vaccines, Curr. Med. Chem. 15, 506–516. 3. Zhong, W., Skwarczynski, M., Fujita, Y., Simerska, P., Good, M. F., and Toth, I. (2009) Design and Synthesis of LipopeptideCarbohydrate Assembled Multivalent Vaccine Candidates Using Native Chemical Ligation, Aust. J. Chem. 62, 993–999. 4. Zhong, W., Skwarczynski, M., Simerska, P., Good, M. F., and Toth, I. (2009) Development of highly pure alpha-helical lipoglycopeptides as self-adjuvanting vaccines, Tetrahedron 65, 3459–3464. 5. Fujita, Y., Moyle, P. M., Hieu, S., Simerska, P., and Toth, I. (2008) Investigation toward multi-epitope vaccine candidates using native chemical ligation, Biopolymers 90, 624–632. 6. Moyle, P. M., Olive, C., Ho, M. F., Good, M. F., and Toth, I. (2006) Synthesis of a highly pure lipid core peptide based self-adjuvanting triepitopic group A Streptococcal vaccine, and subsequent immunological evaluation, J. Med. Chem. 49, 6364–6370. 7. Simerska, P., Abdel-Aal, A. B. M., Fujita, Y., Moyle, P. M., McGeary, R. P., Batzloff, M. R., Olive, C., Good, M. F., and Toth, I. (2008) Development of a liposaccharide-based delivery system and its application to the design of group a streptococcal vaccines, J. Med. Chem. 51, 1447–1452. 8. Moyle, P. M., Olive, C., Karpati, L., Barozzi, N., Ho, M. F., Dyer, J., Sun, H. K., Good, M., and Toth, I. (2005) Synthesis and immunological evaluation of M protein targeted tetra-valent and tri-valent group A

streptococcal vaccine candidates based on the lipid-core peptide system, Int. J. Pept. Res. Ther. 12, 317–326. 9. Hayman, W. A., Toth, I., Flinn, N., Scanlon, M., and Good, M. F. (2002) Enhancing the immunogenicity and modulating the fine epitope recognition of antisera to a helical group A streptococcal peptide vaccine candidate from the M protein using lipid-core peptide technology, Immunol. Cell Biol. 80, 178–187. 10. Abdel-Aal, A. B. M., Batzloff, M. R., Fujita, Y., Barozzi, N., Faria, A., Simerska, P., Moyle, P. M., Good, M. F., and Toth, I. (2008) Structureactivity relationship of a series of synthetic lipopeptide self-adjuvanting group A streptococcal vaccine candidates, J. Med. Chem. 51, 167–172. 11. Olive, C., Hsien, K., Horvath, A., Clair, T., Yarwood, P., Toth, I., and Good, M. F. (2004) Protection against group A streptococcal infection by vaccination with self-adjuvanting lipid core M protein peptides, Vaccine 23, 2298–2303. 12. Olive, C., Batzloff, M. R., Horvath, A., Wong, A., Clair, T., Yarwood, P., Toth, I., and Good, M. F. (2002) A lipid core peptide construct containing a conserved region determinant of the group a streptococcal M protein elicits heterologous opsonic antibodies, Infect. Immun. 70, 2734–2738. 13. Horvath, A., Olive, C., Wong, A., Clair, T., Yarwood, P., Good, M., and Toth, I. (2002) A lipophilic adjuvant carrier system for antigenic peptides, Lett. Pept. Sci. 8, 285–288. 14. Cemazar, M., and Craik, D. J. (2008) Microwave-assisted Boc-solid phase peptide synthesis of cyclic cysteine-rich peptides, J. Pept. Sci. 14, 683–689. 15. Gibbons, W. A., Hughes, R. A., Charalambous, M., Christodoulou, M., Szeto, A., Aulabaugh, A. E.,

308

Skwarczynski and Toth

Mascagni, P., and Toth, I. (1990) Lipidic peptides.1. Synthesis, resolution and structural elucidation of lipidic amino-acids and their homo-oligomers and heterooligomers, Liebigs Ann. Chem., 1175–1183. 16. Schnolzer, M., Alewood, P., Jones, A., Alewood, D., and Kent, S. B. H. (2007) In situ neutralization in boc-chemistry solid phase peptide synthesis - Rapid, high yield assembly of difficult sequences, Int. J. Pept. Res. Ther. 13, 31–44.

17. Schnolzer, M., Alewood, P., Jones, A., Alewood, D., and Kent, S. B. H. (1992) In situ neutralization in boc-chemistry solid phase peptide synthesis - Rapid, high yield assembly of difficult sequences, Int. J. Pept. Protein Res. 40, 180–193. 18. Sarin, V. K., Kent, S. B. H., Tam, J. P., and Merrifield, R. B. (1981) Quantitative monitoring of solid phase peptide-synthesis by the ninhydrin reaction Anal. Biochem. 117, 147–157.

Chapter 19 Coupling Carbohydrates to Proteins for Glycoconjugate Vaccine Development Using a Pentenoyl Group as a Convenient Linker Qianli Wang and Zhongwu Guo Abstract Carbohydrates are important molecular targets in the development of vaccines against cancer, viral and bacterial infections, and many other diseases. However, carbohydrates are usually poorly immunogenic and cannot induce a T cell-dependent immune response that is necessary for effective immunity. To overcome this problem, carbohydrate antigens have to be coupled to an immunogenic carrier molecule, such as a protein, to improve their immunogenicity. To this end, many carbohydrate–protein coupling methods have been developed. A recently established method is based on the introduction of an azido group to carbohydrate antigens during their syntheses, and after the carbohydrate antigens are synthesized, the azido group can be selectively reduced to a free amino group, to which a 4-pentenoyl group can be readily and regiospecifically attached. Thereafter, the C=C bond of the pentenoyl group is ozonolyzed to generate a reactive aldehyde functionality, through which the carbohydrate antigens are linked to carrier proteins by reductive amination. Since the azido group is orthogonal to most transformations involved in carbohydrate synthesis, it can be introduced at an early stage of the synthesis. Moreover, since the pentenoyl group, as well as its aldehyde derivative, is attached to the carbohydrate antigens after they are synthesized, this would significantly simplify the synthetic design of complex carbohydrates, including the design of protecting tactics. Key words: Carbohydrate, Protein, Glycoconjugate, Vaccine, Azide, Pentenoyl group, Reductive amination

1. Introduction The unique carbohydrates expressed by viruses and bacteria or the abnormally and excessively expressed carbohydrates by cancer cells are invaluable molecular targets in the development of vaccines or immunotherapies against the related diseases (1–3). However, to create functional vaccines, carbohydrates must be covalently linked to immunogenic carrier molecules, such as proteins (4, 5), since free carbohydrates are usually poorly immunogenic and fail to Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_19, © Springer Science+Business Media, LLC 2011

309

310

Wang and Guo

induce a T cell-dependent immune response that is necessary for effective immunotherapy (6, 7). For this purpose, many useful carbohydrate–protein coupling methods have been developed (7, 8), among which reductive amination is one of the simplest and the most widely used. In this method, the carbohydrate antigen is modified to contain an aldehyde functionality, which can then react with the free amino groups (usually the side chains of lysine residues) of the carrier protein in the presence of sodium cyanoborohydride (NaBH3CN) to create a stable C–N linkage between the carbohydrate antigen and carrier protein (Fig. 1). To utilize the reductive amination reaction for glycoconjugate vaccine preparation, many linkers containing the aldehyde group or functionalities that can be converted to aldehydes have been developed (3, 9). However, the attachment of these linkers to oligosaccharides at an early stage can make the design of compatible protection–deprotection strategies for the chemical synthesis of oligosaccharides difficult, while late-stage introduction of the linker to the finished oligosaccharides usually requires additional multistep transformations. To circumvent this problem, we have developed a novel synthetic approach, which is based on the installation of an azido group to carbohydrates at the early synthetic stage and then transformation of the azido group to the amide of unsaturated acids followed by the conversion of the unsaturated functionality to an aldehyde by selective oxidation to enable the coupling with proteins by reductive amination (10) (Fig. 2). In this method, the azido group, which is used as a molecular handle for linker attachment, can be installed at the initial stage of carbohydrate synthesis because it is relatively stable toward various transformations

CHO + H2N

n

slow

H2 H C N

NaBH3CN fast

C N n H

glycoconjugate vaccine

carrier protein

carbohydrate

n

Fig. 1. Schematic illustration of the coupling of carbohydrates to proteins by reductive amination. O

R

O

X

R

[H]

N3

O

n

X

NH2

n

carbohydrate antigen X= O, S, N

O3; then Me2S

R

O

X

H N

n

O

2

HO

R 2

O

X

H N

n

O

2

O

Fig. 2. Schematic of the preparation of aldehyde-containing oligosaccharides using an azido group as the molecular handle for the introduction of a pentenoyl group that can be oxidized to form the aldehyde functionality.

Coupling Carbohydrates to Proteins for Glycoconjugate Vaccine Development

311

involved in oligosaccharide synthesis and is a widely adopted functionality as the protected form of amino groups. Ideally, the azido group should be attached to the reducing end of the target carbohydrate antigen since this location is commonly used for protein coupling; otherwise, the carbohydrate reducing end has to be blocked. The introduction of an azido group to the carbohydrate reducing end does not increase the number of synthetic steps; on the contrary, this can actually save several steps of protecting group manipulation normally required. After the synthesis of the carbohydrate antigen is achieved, the azido group can be readily and selectively reduced with phosphites and thiols or by catalytic hydrogenation and so on to form a free amino group suitable for the attachment of an unsaturated acyl group, such as the 4-pentenoyl group. Thereafter, the C=C double bond of the unsaturated acyl group is selectively ozonolyzed to give a carbonyl functionality, affording the aldehyde-containing carbohydrate antigen, which can be coupled with amines through reductive animation as described in Fig.  1. This conjugation method has been successfully employed to construct several carbohydratebased cancer vaccines (10–13). Theoretically, this conjugation method should be applicable to conjugating oligosaccharides to various molecules that contain free amino groups. In this chapter, the conjugation of sTn, a tumor-associated carbohydrate antigen that is richly expressed by a number of tumors (14), to keyhole limpet hemocyanin (KLH), a widely used carrier protein for vaccine development (4), is used as an example (12, 13) to illustrate this new carbohydrate–protein conjugation method (Fig.  3). The experimental protocols are described in detail below. OH HO

OH COOH

OH O

AcHN HO

HO

O

1. H 2 / 10% Pd−C

HO HO AcHN

2.

O O

O

N3

COOH

OH O

AcHN HO

O 2

O 1. O 3

HO HO AcHN

O O

OH HO

O

AcHN HO

OH

COOH

OH

HO KLH

O

HO HO AcHN 3

H N O

2

1

NaBH3CN

O O

H N

CHO O

OH

AcHN HO

2. Me2S

COOH O O HO HO AcHN

O O 4

H N O

N KLH H n

Fig. 3. Schematic of the preparation of an aldehyde-containing sTn antigen derivative and its conjugation with KLH to form an sTn–KLH conjugate.

312

Wang and Guo

2. Materials 2.1. Installation of the Pentenoyl Linker to sTn Antigen

1. sTn antigen derivative 1 (12, 13). 2. 10% Palladium on carbon. 3. Hydrogen gas. 4. 4-Pentenoic anhydride. 5. Biogel P-2 column (2.5 × 30 cm). 6. Dichloromethane (DCM).

2.2. Selective Oxidization of the C =C Bond of the Pentenoyl Linker

1. Pentenoyl linker-containing sTn derivative 2.

2.3. Conjugation of sTn Antigen to KLH

1. Aldehyde-containing sTn derivative 3.

2. Oxone (O3) gas. 3. Dimethyl sulfide (Me2S). 4. Biogel P-2 column (2.5 × 30 cm).

2. Keyhole limpet hemocyanin (KLH). 3. Sodium cyanoborohydride (NaBH3CN). 4. 0.1 M Sodium bicarbonate (NaHCO3) solution, pH 7.5–8.0. 5. 0.1 M Phosphate-buffered saline (PBS), pH 7.8. 6. Biogel A-0.5 m column (1.25 × 30 cm). 7. Bicinchoninic acid (BCA) solution (Sigma, St. Louis, MO). 8. 4% Cupric sulfate (CuSO4) solution: Dissolve 6.25  g of CuSO4·5H2O in 93.75 mL of deionized water (dH2O). 9. Dialysis tubing (6–8 kDa MWCO).

2.4. Analysis of sTn Loading of sTn–KLH Conjugate

1. sTn–KLH conjugate 4. 2. N-Acetyl sialic acid standard solutions: Dissolve 100, 80, 60, 40, 20, 10, 5, 2, 1  mg of N-acetyl sialic acid in 1  mL of dH2O. 3. Resorcinol. 4. 0.1  M CuSO4 solution: Dissolve 2.5  g of CuSO4·5H2O in 100 mL of dH2O. 5. Resorcinol reagent: Dissolve 0.2 g of resorcinol in 80 mL of concentrated HCl, and then add 0.25 mL of 0.1 M CuSO4 solution. The final volume is made up to 100  mL with dH2O. 6. Extraction solvent: n-butyl acetate:n-butanol (85:15, v/v). 7. UV-visible spectrophotometer.

Coupling Carbohydrates to Proteins for Glycoconjugate Vaccine Development

313

3. Methods 3.1. Installation of the Pentenoyl Linker to sTn Antigen

1. Dissolve 32 mg of sTn derivative 1 in 4 mL of dH2O and stir with 10 mg of palladium on carbon catalyst (10% Pd/C, w/w) under a H2 atmosphere at room temperature until the reaction reaches completion (see Note 1). 2. Filter off the catalyst and wash it with 1–2 mL of dH2O. 3. Add 4 mL of MeOH to the filtrate, cool the solution in an ice-water bath, and add 18 mL of 4-pentenoic anhydride (see Note 2). Stir the mixture at room temperature overnight. 4. Concentrate the reaction mixture under vacuum and extract the residue with DCM to remove the nonpolar by-products. 5. Apply the aqueous phase to a Biogel P-2 column (2.5 × 30 cm) using dH2O as eluent. 6. Collect and pool fractions containing the product 2 and lyophilize to give a white solid (~28–32 mg). 7. Confirm the product 2 by nuclear magnetic resonance (NMR) spectroscopy and mass spectrometry (MS).

3.2. Selective Oxidization of the C =C Bond of the Pentenoyl Linker

1. Dissolve 25 mg of 2 in 5 mL of MeOH and cool the solution to −78°C in a dry ice-acetone bath. Bubble ozone into the solution until a blue color appears and remains for 0.5 h (see Note 3). 2. Introduce N2 to the solution to remove the remaining ozone and then add 0.5 mL of Me2S at −78°C. Warm the resultant solution to room temperature over a period of 1 h and stir at room temperature for another 1 h. 3. Condense the reaction mixture under vacuum. 4. Apply the residue to a Biogel P-2 column (2.5 × 30 cm) using dH2O as eluent. 5. Collect and pool fractions containing 3 and lyophilize to give the product as a white solid (~22–24 mg). 6. Confirm the product 3 by NMR and MS spectra (see Note 4).

3.3. Conjugation of sTn Antigen to KLH

1. Dissolve 10 mg of aldehyde 3, 10 mg of KLH, and 10 mg of recrystallized NaBH3CN (see Note 5) in 0.3  mL of 0.1  M NaHCO3, pH 7.5–8.0. 2. Leave the solution to stand in the dark for 3–5  days with occasional shaking. 3. Apply the solution to a Biogel A-0.5 m column (1.25 × 30 cm) using 0.1 M PBS buffer, pH 7.8, as eluent (see Note 6).

314

Wang and Guo

4. Take 10–20  mL of fraction from each tube and add to one well of an enzyme-linked immunosorbent assay (ELISA) microtiter plate. Add 200 mL of BCA solution and 4 mL of 4% CuSO4 solution to each well and incubate the plate at 37°C for 30 min (see Note 7). 5. Dialyze pooled fractions showing a purple color by the above assay against dH2O for 2  days, and lyophilize to yield the expected sTn–KLH conjugate 4 as a white powder (~8–10 mg). 3.4. Analysis of sTn Loading of sTn–KLH Conjugate (4 )

1. Dissolve an exactly weighed amount of sTn–KLH 4 (0.3– 0.5 mg) in 1.0 mL of dH2O in a 16 × 150 mm glass test tube and add 2.0 mL of the resorcinol reagent. 2. Place 1.0 mL of each sialic acid standard solution in a separate 16 × 150  mm glass tube and add 2.0  mL of the resorcinol reagent to each tube. 3. Place the tubes in boiling water for 30 min (see Note 8). 4. Cool the tubes to room temperature. Add 3 mL of extraction solvent into each test tube and vortex the tubes vigorously. Leave the tubes at room temperature for about 10  min to allow the organic solvent layer to separate completely from the aqueous phase. 5. Transfer the top organic phase from each tube to a cuvette and measure the absorbance at 580 nm against pure extraction solvent in an UV-visible spectrophotometer. 6. Create a standard curve using the measured absorbance values of the sialic acid standard solutions. 7. Determine the amount of sialic acid in the sample of sTn–KLH conjugate 4 against the standard curve. 8. Calculate the percentage loading of sTn in the sTn–KLH conjugate 4 according to the following equation (see Note 9):



sTn loading(w / w %) =

amount of sialic acid (mg) in the sample molecular weight of sialic acid ´

molecular weight of sTn weight of sTn − KLH sample (mg)

´ 100%.

4. Notes 1. The reaction can reach completion in about 2–5 h. If in certain cases methanol, instead of water, is used as the solvent for the reaction, it should be treated prior to use with sodium borohydride (NaBH4) to reduce any trace amounts of formaldehyde or acetaldehyde. Otherwise, the resultant amine

Coupling Carbohydrates to Proteins for Glycoconjugate Vaccine Development

315

derived from azido group reduction will react further with formaldehyde or acetaldehyde to form an imine followed by C=N hydrogenation to form methylated or ethylated amines. 2. A large excess of 4-pentenoic anhydride is used to push the reaction to completion quickly and efficiently. The co-solvent methanol ensures that there is only N-acylation but no O-acylation of the substrate, because any excessive 4-pentenoic anhydride would react with methanol rather than with the free hydroxyl group of the substrate. 3. Caution: Ozone has a specific sharp odor resembling chlorine bleach. Exposure to even low concentrations of ozone may cause headaches, burning eyes, and irritation to the respiratory passages. Perform the reaction in a well-ventilated hood! 4. The 1H NMR spectrum of 3 shows a signal corresponding to the hydrated aldehyde at about 5.2  ppm and the complete disappearance of the C=C proton signals. 5. Recrystallization of NaBH3CN is absolutely necessary because commercial NaBH3CN (>98%) contains trace amounts of NaBH4, which quickly reduces aldehydes to form alcohols before participation in the conjugation reaction. NaBH3CN recrystallization procedure: Dissolve 10  g of NaBH3CN in 80 mL of tetrahydrofuran (THF) and add 1 M HCl-methanol until the pH is ~9. Pour the solution with stirring into 250 mL of dioxane and filter the precipitate. Next, dissolve the wet solid in 250  mL of ethyl acetate and filter again. Heat the filtrate to reflux and add (in portions) 150  mL of dioxane with swirling. Slowly cool the solution to room temperature, chill, and filter. Dry the crystals under vacuum at 80°C for 4 h to obtain pure NaBH3CN as a white powder, ~7 g. 6. The conjugate is usually eluted at the void volume and easily separated from the unreacted aldehyde 3. 7. This procedure corresponds to the BCA protein assay method. Protein first chelates with Cu2+ in an alkaline environment containing sodium tartrate, resulting in the reduction of Cu2+ to Cu+. Then, two molecules of BCA chelate with one Cu+ to form a purple-colored reaction product. 8. The tubes may show a range of different colors from blue to purple to brown. The more sialic acid in the solution, the deeper the color is. No blue color means that sialic acid is absent, or the concentration is low. 9. The sTn loading of sTn–KLH conjugate 4 produced by our method is typically 5–10% (w/w). The saccharide loading can reach up to 15% if it is conjugated to a smaller protein such as human serum albumin (HSA) or bovine serum albumin (BSA).

316

Wang and Guo

Acknowledgments Our research on cancer vaccines and cancer immunotherapies is supported by the National Cancer Institute at the National Institutes of Health (CA95142). References 1. Kuberan, B., and Lindhardt, R. J. (2000) Carbohydrate based vaccines, Curr Org Chem 4, 653–677. 2. Ragupathi, G. (1996) Carbohydrate antigens as targets for active specific immunotherapy, Cancer Immunol Immunother 43, 152–157. 3. Danishefsky, S. J., and Allen, J. R. (2000) From the laboratory to the clinic: A retrospective on fully synthetic carbohydrate-based anticancer vaccines, Angew Chem, Int Ed 39, 836–863. 4. Helling, F., Shang, A., Calves, M., Zhang, S., Ren, S., Yu, R. K., Oettgen, H. F., and Livingston, P. O. (1994) GD3 vaccines for melanoma: superior immunogenicity of keyhole limpet hemocyanin conjugate vaccines, Cancer Res 54, 197–203. 5. Livingston, P. O. (1995) Approaches to augmenting the immunogenicity of melanoma gangliosides: from whole melanoma cells to ganglioside-KLH conjugate vaccines, Immunol Rev 145, 147–166. 6. Chen, C. H., and Wu, T. C. (1998) Experimental vaccine strategies for cancer immunotherapy, J Biomed Sci 5, 231–252. 7. Jennings, H. J., and Sood, R. K. (1994) Synthetic glycoconjugates as human vaccines, in Neoglycoconjugates: Preparation and Applications (Lee, Y. C., and Lee, R. T., Eds.), pp 325–371, Academic Press, San Diego. 8. Zou, W., and Jennings, H. J. (2009) Preparation of glycoconjugate vaccines, in CarbohydrateBased Vaccines and Immunotherapies (Guo, Z., and Boons, G.-J., Eds.), pp 263–312, John Wiley & Sons, Hoboken.

9. Allen, J. R., and Danishefsky, S. J. (1999) New applications of the n-pentenyl glycoside method in the synthesis and immunoconjugation of fucosyl GM(1): A highly tumorspecific antigen associated with small cell lung  carcinoma, J Am Chem Soc 121, 10875–10882. 10. Xue, J., Pan, Y., and Guo, Z. (2002) Neoglycoprotein cancer vaccines: Synthesis of an azido derivative of GM3 and its efficient coupling to proteins through a new linker, Tetrahedron Lett. 43, 1599–1602. 11. Pan, Y. B., Chefalo, P., Nagy, N., Harding, C., and Guo, Z. W. (2005) Synthesis and immunological properties of N-modified GM3 antigens as therapeutic cancer vaccines, J Med Chem 48, 875–883. 12. Wang, Q. L., Ekanayaka, S. A., Wu, J., Zhang, J. P., and Guo, Z. W. (2008) Synthetic and Immunological Studies of 5¢-N-Phenylacetyl sTn to Develop Carbohydrate-Based Cancer Vaccines and to Explore the Impacts of Linkage between Carbohydrate Antigens and Carrier Proteins, Bioconjugate Chem 19, 2060–2067. 13. Wu, J., and Guo, Z. (2006) Improving the antigenicity of sTn antigen by modification of its sialic acid residue for development of glycoconjugate cancer vaccines, Bioconjug Chem 17, 1537–1544. 14. Slovin, S. F., Keding, S. J., and Ragupathi, G. (2005) Carbohydrate vaccines as immunotherapy for cancer, Immunol Cell Biol 83, 418–428.

Chapter 20 Conjugation of LPS-Derived Oligosaccharides to Proteins Using Oxime Chemistry Joanna Kubler-Kielb Abstract Conjugates of bacterial polysaccharides covalently bound to a carrier protein are among licensed human vaccines. Immunization of adults and children with these vaccines results in induction of saccharidespecific antibodies composed mainly of the IgG class. Depending on the choice of coupling technique, saccharides can be attached to a protein by either multiple- or single-point attachments. While the first method is suitable for high molecular mass polysaccharides, the second one is beneficial for low-molecular mass compounds such as synthetic carbohydrates or bacterial oligosaccharides obtained by different degradation procedures. This chapter describes a method for coupling low-molecular mass lipopolysaccharide (LPS)-derived oligosaccharides composed of a core or a short O-specific polysaccharide-core fragment (O-SPC) to a carrier protein by a single-point attachment. Conjugation is performed between the carbonyl group of the reducing terminal of 3-deoxy-d-manno-oct-2-ulosonic acid (Kdo) exposed after acid hydrolyses of LPS and the aminooxy group of a bifunctional linker bound to the protein. This is an efficient reaction that can be carried out quickly and under mild conditions. Conjugates thus prepared using this approach preserve the external nonreducing end of the sugar chain and can induce antibodies to both conjugate components. Consequently, this method is highly suitable for the preparation of LPSbased human vaccines. Key words: LPS, Vaccine, Conjugate, Kdo, Aminooxy, Hydroxylamine, Oxime

1. Introduction Conjugations of saccharides (T-cell independent antigens) to carrier proteins (T-cell dependent antigens) were first introduced in 1980’ using Haemophilus influenzae type b capsular polysaccharide. Conjugation both increased saccharide immunogenicity and facilitated T-cell dependent properties, enabling booster responses and providing vaccines for infants (1, 2). These vaccines eliminated most of the meningitis cases caused by this organism.

Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_20, © Springer Science+Business Media, LLC 2011

317

318

Kubler-Kielb

This finding led to a rapid development of conjugate vaccines against other prevalent pathogens such as Neissieria meningitidis, Streptococcus pneumoniae, and Salmonella typhi (3). Currently, commercially available vaccines are composed of bacterial capsular polysaccharides (CPS) bound to a protein carrier mostly by multiplepoint attachments (i.e., a “lattice”-type configuration). In nonencapsulated Gram-negative bacteria, the outermost exposed carbohydrate antigens are their lipopolysaccharides (LPS), which may be present in three forms: (1) Lipid A and a core oligosaccharide only; (2) Lipid A, a core, and a small number (1–5) of O-specific polysaccharide (O-SP) repeat units (RUs); and (3) Lipid A, a core, and a larger number (20–25) of O-SP RUs. LPS, however, differs from CPS: LPS is toxic and therefore it must be detoxified before use by the removal of the lipid component. Upon removal of the lipid domain, the molecular mass of the remaining carbohydrate is much lower than that of most CPS molecules. Investigational vaccines composed of “lattice”-type protein conjugates of the O-SP of Shigella dysenteriae type 1, Shigella sonnei, and Shigella flexneri 2a have been shown to be safe, and elicited specific LPS antibodies in adults and children in several studies (4–6). Furthermore, a S. sonnei O-SP conjugate showed >70% efficacy in Israeli army recruits and in children over 3 years of age (7, 8). And finally, other investigators have reported that short synthetic S. dysenteriae type 1 saccharides containing 2, 3 or 4 O-SP RUs and low-mass S. sonnei O-specific polysaccharide-core (O-SPC) fragments containing the core domain plus an average of 3–4 O-SP RUs bound to carrier proteins by a single-point attachment (i.e., a “sun”-type configuration) were more immunogenic in mice than the O-SP conjugates tested clinically (9, 10). Therefore, an efficient method for conjugating LPS-derived short-length oligosaccharides to proteins may find wide application in the preparation of vaccines against a broad range of Gram-negative pathogens. Numerous methods have recently been proposed for synthesizing “sun”-type vaccine conjugates, with reductive amination being the most popular (11–13). The choice of a suitable coupling protocol is restricted by the pH and temperature sensitivity of the vaccine components to be conjugated. It should further be noted that the types of chemical linkages formed between the saccharide and the protein, the saccharide chain length, the chain density on the protein and the non-reducing end moiety may influence serum antibody responses to both vaccine components. To avoid modification of the sugar structure and/or protein denaturation, the coupling procedure should be performed under mild conditions near neutral pH. Moreover, the linking reagent should react at the desired site only, the intermediates should be stable during purification steps, and the procedure should allow the recovery of the unconjugated material in its original chemical form. And finally, the linker should be appropriate for human use. With these goals in mind, we developed a conjugation

Conjugation of LPS-Derived Oligosaccharides to Proteins Using Oxime Chemistry

319

­ rotocol for binding LPS-derived O-SPC or core saccharides to p the aminooxy group of a bifunctional linker bound to a carrier protein (10, 14).

2. Materials 2.1. Preparation of Oligosaccharides from LPS

1. Lipopolysaccharide (LPS) is obtained from commercial sources or isolated by hot phenol-water extraction from ­bacterial cells, and purified by enzyme treatment and ultracentrifugation as described in ref. (15) (see Note 1). 2. Buffer A: Pyridine/concentrated acetic acid/water (4/8/988v/ v/v, pH 4.5). 3. Bio-Gel P-4, Bio-Gel P-10 gel filtration columns (1 × 100 cm) (Bio-Rad, Hercules, CA) equilibrated in Buffer A. 4. Differential refractometer (Knauer, Berlin, Germany).

2.2. Derivatization of Proteins with an Aminooxy Group

1. Bovine serum albumin (BSA). 2. Succinimidyl 3-(bromoacetamido)propionate (SBAP) (Pierce, Rockford, IL). 3. O-3-(thiopropyl)hydroxylamine hydrochloride can be synthesized using the procedure described in ref. (16), or the aminooxy linker can also be obtained from commerical suppliers (e.g., custom synthesized by SoluLink, San Diego, CA). 4. Dimethyl sulfoxide (DMSO). 5. 0.1 M NaOH. 6. Buffer B (10× concentrated solution): Dissove 15.05  g anhydrous K2HPO4, 1.85  g KH2PO4, 5.84  g NaCl, 4.16  g ethylenediaminetetraacetic acid tetrasodium salt dihydrate (EDTA), and 1  ml glycerol in water and bring the final volume to 100 ml. Adjust the pH of the solution to 7.4. 7. Sephadex® G-50 gel filtration column (1 × 50  cm) (GE Healthcare, Piscataway, NJ) equlibrated in 0.2 M NaCl. 8. Amicon® Ultra Centrifugal Filter Devices, Ultracel PL-10 (10,000 MWCO), 15 mL (Millipore, Cork, Ireland).

2.3. Conjugation

1. Sephadex® G-75 gel filtration column (1 × 100  cm) (GE Healthcare) equilibrated in 0.2 M NaCl.

2.4. Characterization of Derivatized Proteins and Conjugates

1. BCA Protein Assay Kit (Thermo Scientific, Rockford, IL).

2.4.1. Protein Colorimetric Assays

320

Kubler-Kielb

2.4.2. Sugar Colorimetric Assays

1. Anthrone (J.T. Baker, Phillipsburg, NJ). 2. Concentrated sulfuric acid. 3. Ice water bath. 4. UV-visible spectophotometer.

2.4.3. SDS-PAGE Analysis

1. NuPAGE® 10% Bis-Tris gel (1.0 × 10 mm well) (Invitrogen, Carlsbad, CA). 2. NuPAGE® LDS Sample Buffer (4×) (Invitrogen). 3. NuPAGE® MOPS SDS Running Buffer (Invitrogen).

2.4.4. Double Immunodiffusion

1. Agarose (Invitrogen). 2. Anti-BSA antibody (Sigma-Aldrich, St. Louis, MO). 3. Anti-LPS precipitating sera.

2.4.5. Matrix-Assisted Laser Desorption/Ionization Time-of-Flight Mass Spectrometry (MALDI-TOF MS) Analysis

1. MALDI-TOF MS matrix is prepared by dissolving sinapinic acid (Sigma-Aldrich) in a solution containing 30% CH3CN/0.1% trifluoroacetic acid (v/v in water).

3. Methods The scheme for preparing conjugates by derivatization of protein with O-(3-thiopropyl)hydroxylamine (two-step procedure), followed by reaction with a carbonyl group of acetic acid-hydrolyzed 3-deoxy-d-manno-oct-2-ulosonic acid (Kdo) located at the reducing end of core saccharides or O-SP-core fragments to form a stable oxime linkage, is outlined in Fig. 1. 3.1. Preparation of Oligosaccharides from LPS

1. LPS (100  mg) is heated in 10  ml of water containing 1% (v/v) acetic acid for 90 min at 100°C. 2. The mixture is ultracentrifuged at 142,000 × g for 5 h at 4°C. 3. The carbohydrate-containing supernatant is freeze-dried, and then dissolved in 1–2 ml of Buffer A. 4. The solution is passed through a Bio-Gel P-4 or P-10 gel filtration column (1 × 100  cm) (see Note 2) in Buffer A, and monitored with a Knauer differential refractometer (10, 14). An example of the separation of S. sonnei O-SPC using a BioGel P-10 gel filtration column is presented in Fig. 2.

3.2. Derivatization of Proteins with an Aminooxy Group

BSA (see Note 3) is aminooxylated in a two-step procedure as follows.

321

Conjugation of LPS-Derived Oligosaccharides to Proteins Using Oxime Chemistry Pr-NH2

O

O

O N

PBS, pH 7.4 o

2h, 23 C

Br

N H

O

O

Succinimidyl 3-(bromoacetamido) propionate (SBAP) Pr

PBS, pH 7.4

HS

2h, 23oC

O-SP-core-Kdo-LipidA

Br

N H

N H

LPS

O

O

O

1% acetic acid 1h, 100oC

NH2

O-(3-thiopropyl)hydroxylamine O

O Pr

N H

O-SP-core-Kdo

O

S

N H

NH2

PBS, pH 5.5, 16h, 23oC

O

O Pr

N H

N H

S

O

N

Kdo-core-O-SP

Fig. 1. Schematic outline of the conjugation of O-SPC oligosaccharides to protein. Protein is first aminooxylated in a twostep procedure and reacted with the carbonyl group of the reducing terminal of 3-deoxy-d-manno-oct-2-ulosonic acid (Kdo) exposed after acid hydrolyses of LPS. The external non-reducing end of the O-SPC is preserved in these conjugates.

a

b O-SPC av. 3-4 RU+core

Refraction Index

O-SP

O-SPC av. 1-2 RU+core O-SP

O-SPC 35

45

55

65

75

85

95

105

115

Effluent [ml]

Fig. 2. (a) SDS-PAGE analysis of S. sonnei LPS showing the presence of O-SP and O-SPC fragments. (b) Bio-Gel P-10 gel filtration of S. sonnei LPS hydrolyzed in 1% (v/v) acetic acid in water.

322

Kubler-Kielb

3.2.1. Synthesis of Bromoacetyleted BSA (BSA-Br)

1. A solution of BSA (90 mg, 1.4 mmol containing 81 mmol of lysine) in 1.5 ml Buffer B (pH 7.4) is treated with succinimidyl 3-(bromoacetamido)propionate (SBAP) (20  mg, 65  mmol) dissolved in 50 ml of DMSO, at 23°C, pH 7.4; the pH of the mixture is maintained at 7.4 by the addition of 0.1 M NaOH. 2. The derivatization reaction with SBAP proceeds most rapidly within the first 20 min. After 2 h, the BSA/SBAP mixture is applied to a Sephadex® G-50 gel filtration column (1 × 50 cm) equilibrated in 0.2 M NaCl (see Note 4). 3. The void volume fractions are pooled and concentrated by Amicon® Ultra Centrifugal Filters to ~2 ml. 4. The purified solution containing bromoacetylated BSA (BSA-Br) is analyzed for protein concentration, antigenicity by double immunodiffusion and molecular mass by SDSPAGE (Fig.  3) and MALDI-TOF mass spectrometry (see Subheading 3.4). The observed average molecular mass of bromoacetyleted BSA is typically ~72 kDa, indicating the incorporation of an average of 30 SBAP linker molecules (see Note 5). The average yield of this step is ~85% as determined by protein concentration assays.

Fig. 3. SDS-PAGE analysis of BSA/B. pertussis core conjugates with different saccharide chain densities per protein molecule. Lane 1: BSA, Lane 2: BSA-Br, Lane 3: BSA-ONH2, and Lane 4: BSA/B. pertussis core conjugate (20 chains of saccharide chains per protein molecule; conjugation was performed using 1:1 (w/w) ratio of BSA-ONH2 to saccharide), Lane 5: BSA/B. pertussis core conjugate (12 chains of saccharide chains per protein molecule; conjugation was performed using 1:0.5 (w/w) ratio of BSA-ONH2 to saccharide).

Conjugation of LPS-Derived Oligosaccharides to Proteins Using Oxime Chemistry 3.2.2. Synthesis of Aminooxylated BSA

323

1. BSA-Br in 1.5 ml Buffer B (70 mg, ~29 mmol of bromoacetyl groups) is reacted with O-3-(thiopropyl)hydroxylamine hydrochloride (12 mg, 84 mmol, dissolved in 200 ml of 1 M K2 HPO4) for 2  h at 23°C, pH 7.4; the pH of the mixture is maintained by addition of 0.1 M NaOH. 2. The mixture is then applied to a gel filtration column as described in Subheading 3.2.1 above. 3. The void volume fractions are pooled and concentrated by Amicon Ultra Centrifugal Filters to ~2 ml. 4. The purified solution containing aminooxylated BSA (BSAONH2) is analyzed for protein concentration, antigenicity by double immunodiffusion and molecular mass by SDS-PAGE (Fig.  3) and MALDI-TOF mass spectrometry (Fig.  4) (see Subheading 3.4). The average molecular mass of aminooxylated BSA is typically ~73 kDa, and the average yield of this step is ~80% as determined by protein assays.

3.3. Conjugation

1. BSA-ONH2, 5 mg, is reacted with 5 mg of O-SPC or core saccharide in 1.5 ml Buffer B (adjusted to pH 5.5 with 0.05 M HCl) for 15 h with stirring at 23°C (see Note 6). 2. The reaction mixture is passed through a Sephadex® G-75 gel filtration column (1 × 100  cm) equilibrated in 0.2  M NaCl (see Note 7). 3. The void volume fractions are pooled and characterized for protein and sugar concentrations, antigenicity by double immunodiffusion (Fig. 5) and molecular mass by SDS-PAGE (Fig.  3) and MALDI-TOF mass spectrometry (Fig.  4) as described in Subheading 3.4 below.

Fig.  4. Examples of MALDI-TOF spectra used for calculation of the average number of  aminooxy linkers and oligosaccharide chains incorporated per protein molecule. (a) BSA-ONH2. (b) BSA/S. sonnei O-SPC conjugate (Molecular mass values are in Da).

324

Kubler-Kielb

Fig. 5. Double immunodiffusion analysis of BSA/S. dysenteriae type 1 O-SPC conjugate (Well C ) with anti-S. dysenteriae type 1 (Well S ) and anti-BSA (Well B ) sera. A line of identity between the conjugate sample and the two antisera is clearly visible.

3.4. Characterization of Derivatized Proteins and Conjugates

Protein concentration is measured by using the BCA Protein Assay Kit according to the manufacturer’s instructions.

3.4.1. Protein Concentration Determination 3.4.2. Sugar Concentration Analysis

Sugar concentration is measured by the Anthon assay (17) as follows: 1. Reagent 1 is prepared by dissolving 50  mg of anthrone in 25 ml of concentrated sulfuric acid, 4 h before use. 2. The sample (0.5 ml) is mixed with 1 ml of Reagent 1, heated at 100°C for 10 min, and then cooled down in an ice water bath. 3. The absorbance of the solution is read at 620  nm using a UV-visible spectrophotometer. 4. The sugar concentration is determined by comparing the measured OD against a standard curve generated for serial dilutions of the oligosaccharide used for conjugation (usually 10–100 mg/sample).

3.4.3. Double Immunodiffusion Analysis

1. 0.9 g of agarose is dissolved in 100 ml of PBS by boiling. 2. 10 ml of the heated solution (~50°C) is poured onto 2 × 3-in. glass slides and the plates are cooled down to ~3–6°C. Small wells (holes) for loading samples are punched into the gel plates. The wells are spaced ~0.5- to 0.75-cm apart from one another. 3. Protein samples are loaded into the gel plate wells and reacted against anti-protein and anti-LPS sera for 12–48 h at room temperature. A line of identity between the antigen and the two antisera is expected. An example of a double immunodiffusion analysis of a BSA/S. dysenteriae type 1 O-SPC conjugate is presented in Fig. 5.

Conjugation of LPS-Derived Oligosaccharides to Proteins Using Oxime Chemistry

325

3.4.4. SDS-PAGE Analysis

10% NuPage® gels are run according to the manufacturer’s instructions using MOPS running buffer; 5–10 mg of protein are loaded per well.

3.4.5. MALDI-TOF Analysis

1. Derivatized protein or conjugates obtained after gel filtration are concentrated to ~10–20  mg/ml in 0.2  M NaCl (see Note 7). 2. 1 ml of the sample is mixed with 20 ml of sinapinic acid matrix prepared in 30% CH3CN and 0.1% trifluoroacetic acid (v/v in water). 3. 1 ml of the mixture is applied onto the sample stage, dried, and placed in the mass spectrometer for analysis.

4. Notes 1. The final product should contain <5% contamination from nucleic acids or proteins. The nucleic acid content is assessed by measuring absorbance using a UV-visible spectrophotometer; 1.0 OD260 = 50 mg of nucleic acids in the sample. 2. The choice of gel filtration column to employ depends on the type of LPS used for conjugation. For long LPS molecules, O-SPC fractions containing the core domain plus several RUs of O-SP are separated from the full length O-SP using a BioGel P-10 column as described for Shigella sonnei (10). An example column chromatography profile is presented in Fig. 2. If the LPS has only the core domain without any O-SP component, as in the case of Bordetella pertussis or Haemophilus ducreyi, a Bio-Gel P-4 column is recommended; in this case, the void volume fractions are collected. However, the most suitable choice of gel column has to be determined for each LPS tested. Isolated oligosaccharide fractions should be analyzed by nuclear magnetic resonance (NMR) spectroscopy and mass spectrometry to confirm their structures and lengths (10, 14). 3. This example protocol describes the derivatization of BSA; however, the method can be adapted for use with other proteins of interest. 4. Proteins sensitive to pH may require the use of an alternative elution buffer such as PBS or Buffer B. 5. Depending on the protein used, the efficiency of incorporation of linkers will vary (mainly because of the amount and the availability of lysine amino acid residues). 6. The weight proportion of the amounts of protein and sugar to use will vary depending of the type of LPS used and on the

326

Kubler-Kielb

desired number of sugar chains incorporated per protein. An example of the incorporation of different densities of sugar chains per protein obtained by using different amounts of sugar in the reaction mixture is presented in Fig. 3 for BSA/B. pertussis core conjugates. In general, LPS molecules having phosphorylated Kdo (Bordetella, Haemophilus spp.) are more reactive with the aminooxy linker than LPS molecules in which Kdo is substituted by another Kdo (Enterobacteriaceae); in general, the longer the oligosaccharide, the fewer chains will be coupled per protein. Furthermore, the pH of the reaction has to be adjusted to a level at which the protein and the saccharide can tolerate without precipitation or structural modifications. It is recommended that the reaction mixture be maintained between pH 5.5 and 7.5; the higher the pH of the solution, the longer the time needed for efficient incorporation of oligosaccharides. 7. For some proteins, it maybe necessary to replace the elution buffer with water; this may be accomplished washing the sample four to five times using an Amicon® Ultra Centrifugal Filter device.

Acknowledgments This research was supported by the Intramural Research Program of the NICHD at the National Institutes of Health. The author thanks Drs. Rachel Schneerson and John B. Robbins for critical review of this manuscript. References 1. Schneerson, R., Barrera, O., Sutton, A., and Robbins, J.B. (1980) Preparation, characterization, and immunogenicity of Haemophilus influenzae type b polysaccharide-protein ­conjugates. J. Exp. Med. 152, 361–76. 2. Robbins, J. B., and Schneerson, R. (1990) Polysacchride protein conjugates: A new ­generation of vaccines. J. Infect. Dis. 161, 821–32. 3. Pon, R. A., and Jennings, H. J. (2009) Carbohydrate-based antibacterial vaccines. In. Carbohydrate-Based Vaccines and Immuno­ therapies (Guo, Z., and Boons, G. J., eds.), John Wiley & Sons, New York, pp. 117–166. 4. Taylor, D. N., Trofa, A.C., Sadoff, J., Chu, C., Bryla, D., Shiloach, J., Cohen, D., Ashkenazi, S., Lerman, Y., Egan, W., Schneerson, R. and Robbins, J. B. (1993) Synthesis, characterization,

and clinical evaluation of conjugate ­vaccines composed of the O-specific polysaccharides of Shigella dysenteriae type 1, Shigella flexneri type 2a, and Shigella sonnei (Plesiomonas shigelloides) bound to bacterial toxoids. Infect. Immun. 61, 3678–87. 5. Passwell, J. H., Harlev, E., Ashkenazi, S., Chu, C., Miron, D., Ramon, R., Farzan, N., Shiloach, J., Bryla, D. A, Majadly, F., Roberso,n R., Robbins, J. B., and Schneerson, R. (2001) Safety and immunogenicity of improved Shigella O-specific polysaccharide-protein conjugate vaccines in adults in Israel. Infect. Immun. 69, 1351–7. 6. Passwell, J. H., Ashkenazi , S., Harlev, E., Miron, D., Ramon, R., Farzam, N., LernerGeva, L., Levi, Y., Chu, C., Shiloach, J., Robbins, J. B., and Schneerson, R. (2003)

Conjugation of LPS-Derived Oligosaccharides to Proteins Using Oxime Chemistry Safety and immunogenicity of Shigella sonneiCRM9 and Shigella flexneri type 2a-rEPAsucc conjugate vaccines in one- to four-year-old children. Pediatr. Infect. Dis. J. 22, 701–6. 7. Cohen, D., Ashkenazi, S., Green, M. S., Gdalevich, M., Robin, G., Slepon, R., Yavzori, M., Orr, N., Block, C., Ashkenazi, I., Shemer, J., Taylor, D. N., Hale, T. L., Sadoff, J. C., Pavliakova, D., Schneerson, R., and Robbins, J. B. (1997) Double-blinded vaccine-controlled ­randomized efficacy trial of an investigational Shigella sonnei conjugate vaccine in young adults. Lancet 349, 155–9. 8. Passwell, J. H., Ashkenazi, S., Levi, Y., BanetLevi, Y., Ramon-Saraf, R., Farzam, N., LernerGeva, L., Even-Nir, H., Chu, C., Shiloach, J., Robbins, J. B., and Schneerson, R. (2010) Age-related efficacy of Shigella O-specific polysaccharide conjugates in 1–4-year-old Israeli children. Vaccine 28, 2231–2235. 9. Pozsgay, V., Chu, C., Pannell, L., Wolfe, J., Robbins, J. B., and Schneerson, R. (1999) Protein conjugates of synthetic saccharides elicit higher levels of serum IgG lipopolysaccharide antibodies in mice than do those of the O-specific polysaccharide from Shigella dysenteriae type 1. Proc. Natl. Acad. Sci. USA 96, 5194–57. 10. Robbins, J. B., Kubler-Kielb, J., Vinogradov, E., Mocca, C., Pozsgay, V., Shiloach, J., and Schneerson, R. (2009) Synthesis, characterization, and immunogenicity in mice of Shigella sonnei O-specific oligosaccharid­e-core-protein conjugates. Proc. Natl. Acad. Sci. USA 106, 7974–8.

327

11. Pozsgay, V., and Kubler-Kielb, J. (2008) Conjugation methods towards synthetic vaccines. In ACS Symposium Series Volume 989: Carbohydrate-Based Vaccines (Roy, R., ed.), American Chemical Society, Washington D.C., pp. 36–70. 12. Zou, W., and Jennings, H. J. (2009) Preparation of glycoconjugate vaccines. In CarbohydrateBased Vaccines and Immunotherapies (Guo, Z., and Boons, G. J., eds.), John Wiley & Sons, New York, pp. 55–88. 13. Lees, A., Sen, G., and LopezAcosta, A. (2006) Versatile and efficient synthesis of protein-polysaccharide conjugate vaccines using aminooxy reagents and oxime chemistry. Vaccine 24, 716–29. 14. Kubler-Kielb, J., Vinogradov, E., BenMenachem, G., Pozsgay, V., Robbins, J. B., and Schneerson, R. (2008) Saccharide/protein conjugate vaccines for Bordetella species: preparation of saccharide, development of new conjugation procedures, and physico-chemical and immunological characterization of the conjugates. Vaccine 26, 3587–93. 15. Westphal, O., Jann, K., and Himmelspach, K. (1983) Chemistry and immunochemistry of bacterial lipopolysaccharides as cell wall antigens and endotoxins. Prog. Allergy 33, 9–39. 16. Kubler-Kielb, J., and Pozsgay, V. (2005) A new method for conjugation of carbohydrates to proteins using an aminooxy-thiol heterobifunctional linker. J. Org. Chem. 70, 6987–90. 17. Bailey, R. W. (1958) The reaction of pentoses with anthrone. Biochem. J. 68, 669–672.

Chapter 21 Site-Specific Chemical Modification of a Glycoprotein Fragment Expressed in Yeast Junpeng Xiao and Thomas J. Tolbert Abstract Site-specific modification of glycoproteins has wide application in both biochemical and biophysical studies. This method describes the conjugation of synthetic molecules to the N-terminus of a glycoprotein fragment, viz., human immunoglobulin G subclass 1 fragment crystallizable (IgG1 Fc), by native chemical ligation. The glycosylated IgG1 Fc is expressed in a glycosylation-deficient yeast strain. The N-terminal cysteine is generated by the endogenous yeast protease Kex2 in the yeast secretory pathway. The N-terminal cysteine is then conjugated with a biotin thioester to produce a biotinylated, glycosylated IgG1 Fc using native chemical ligation. Key words: Glycoprotein, IgG1 Fc, Site-specific chemical modification, Native chemical ligation, N-terminal cysteine, Biotin

1. Introduction The site-specific chemical modification of proteins has wide application, including the incorporation of labels, probes and posttranslational modifications for biochemical and biophysical studies, and the incorporation of synthetic modifications within proteins to alter their bioactivity and physical properties (1–7). N-terminal cysteines have been widely used for site-specific protein N-terminal modification using chemoselective ligations, such as native chemical ligation (NCL) between N-terminal cysteines and thioesters (8–14), thiazolidine ligation between N-terminal cysteines and aldehydes (15, 16), and condensation between N-terminal cysteines and cyanobenzothiazole derivatives (17). However, the production of expressed glycoproteins with

Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_21, © Springer Science+Business Media, LLC 2011

329

330

Xiao and Tolbert

N-terminal cysteines for site-specific glycoprotein modification has only recently been reported by our laboratory (7). Here, we present a detailed method for the production of N-terminal cysteine-containing glycoproteins in yeast for site-specific chemical modification. In this method, a model glycoprotein, human immunoglobulin G subclass 1 fragment crystallizable (IgG1 Fc) containing an N-terminal cysteine, is expressed in a glycosylation-deficient yeast strain. The N-terminal cysteine is generated by proteolytic removal of a propeptide leader sequence by the endogenous yeast protease Kex2 in the yeast secretory pathway. Next, the N-terminal cysteine is conjugated with a synthetic biotin thioester to produce a biotinylated, glycosylated IgG1 Fc using native chemical ligation as an example of glycoprotein modification. The biotinylated glycoproteins produced by this method can be used for western-blotting or site-specific protein immobilization; moreover, these methods are compatible for use with a wide variety of thioester labels and other synthetic modifications (18, 19).

2. Materials 2.1. Chemical Synthesis of the Biotin Thioester

1. Potassium ethylxanthate. 2. Acetone. 3. tert-Butyl chloroacetate. 4. Diethyl ether. 5. Ethanolamine. 6. Ethyl acetate. 7. Biotin. 8. Dimethylformamide (DMF). 9. 1,3-Diisopropylcarbodiimide (DIC). 10. Methanol. 11. Dichloromethane (DCM). 12. Trifluoroacetic acid (TFA). 13. Toluene. 14. Celite®. 15. Rotary evaporator. 16. Nuclear magnetic resonance (NMR) spectrometer. 17. Matrix-assisted laser desorption/ionization Fourier transform (MALDI-FT) mass spectrometer.

Site-Specific Chemical Modification of a Glycoprotein Fragment Expressed in Yeast

2.2. Construction of the pGAP-C-IgG1 Fc Expression Plasmid

331

1. pGAPZaA plasmid (Invitrogen). Store at −20°C. 2. Restriction enzymes: Xho I and Not I (New England Biolabs Inc.). Store at −20°C. 3. Antarctic phosphatase (New England Biolabs Inc.). Store at −20°C. 4. PCR Primers: 5¢-XhoI-CTC-IgG Fc (5¢-GGCCCGCTCGAGAAAAGATGC ACATGCCCACCGTGCCCAGCA-3¢) 3¢-NotI-stop-IgG Fc (5¢-GGGCCCGCGGCGGCCGCTCA TTTACCCGGAGACAGGGAGAG-3¢) Store the primers at −20°C. 5. MGC-12853 plasmid (Mammalian Gene Collection) (20). Store at −20°C. 6. T4 DNA ligase (New England Biolabs Inc.). Store at −20°C. 7. TOP10F’ chemically competent (Invitrogen). Store at −80°C.

Escherichia coli

cells

8. Zeocin (100 mg/mL in water) (Invitrogen). Store at −20°C. 9. Low-salt LB/Zeocin plates: 1% (w/v) peptone, 0.5% (w/v) yeast extract, 0.5% (w/v) NaCl, 1.5% (w/v) agar, and 25  mg/mL Zeocin. 2.3. Expression and Purification of Glycosylated C-IgG1 Fc

1. OCH1-deleted strain of Pichia pastoris SMD1168. This ­glycosylation-deficient yeast strain was produced in our laboratory using previously published methods (7, 21). 2. Restriction enzymes: Avr II (New England Biolabs Inc.). Store at −20°C. 3. YPD/Zeocin plates: 1% (w/v) yeast extract, 2% (w/v) peptone, and 2% (w/v) glucose, 2%(w/v) agar, and 100 mg/mL Zeocin. 4. YPD/Zeocin medium: 1% (w/v) yeast extract, 2% (w/v) peptone, and 2% (w/v) glucose, and 100 mg/mL Zeocin. 5. Supplemented YPD medium: 1% (w/v) yeast extract, 2% (w/v) peptone, and 2% (w/v) glucose, 4 × 10−5% (w/v) ­biotin, 0.004% (w/v) histidine, 1.34% (w/v) yeast nitrogen base with ammonium sulfate without amino acids, and 100 mM potassium phosphate buffer, pH 6. 6. Antifoam 204 (Sigma), sterilized by autoclaving. 7. 20% (w/v) glucose stock solution, sterilized by autoclaving. 8. Protein G Sepharose®, Fast Flow chromatography medium (Sigma-Aldrich). 9. Phosphate buffer: 20 mM sodium phosphate, pH 7.0.

332

Xiao and Tolbert

10. Tris buffer: 1 M Tris–HCl base, pH 7.8. 11. Elution buffer: 100 mM glycine, pH 2.7. 12. Phenyl Sepharose® Aldrich).

chromatography

medium

(Sigma-

13. Buffer A: 20 mM sodium phosphate, 1 M ammonium sulfate, pH 7.0. 14. Dithiothreitol, DTT. 15. Acetonitrile (ACN). 2.4. Ligation of the Biotin Thioester to the C-IgG1 Fc

1. Hydroxylamine buffer: 20 mM sodium phosphate, 200 mM hydroxylamine, pH 5.0. 2. Ligation buffer: 20 mM sodium phosphate, 5 mM betaine, pH 7.5. 3. Sodium 2-mercaptoethanesulfonate.

2.5. Deglycosylation of Biotin-Modified Glycosylated C-IgG1 Fc

1. PNGase F glycosidase enzyme, 500,000  units/ml (New England Biolabs Inc.). Stored at −20°C.

3. Methods This method describes the conjugation of biotin to the N-terminus of an IgG1 Fc glycoprotein using native chemical ligation (NCL), which is a chemoselective ligation reaction that joins N-terminal cysteine-containing peptides and proteins to thioester-containing molecules (8). In this method, the N-terminal cysteine-containing glycosylated IgG1 Fc is produced by expression in a glycosylationdeficient yeast strain, and then labeled with a biotin thioester (6) that is chemically synthesized as described below. 3.1. Chemical Synthesis of the Biotin Thioester, 2-(5-(2oxohexahydro­thieno(3,4-d) imidazol-4-yl) pentanoylthio)acetic acid6 (6)

The carboxylic acid of biotin is coupled to tert-butyl 2-mercaptoacetate (22) using DIC. The tert-butyl protecting group is then removed by TFA to form a water-soluble biotin thioester (Fig. 1) (see Note 1). 1. Potassium ethylxanthate 1 (5.0 g, 31.2 mmol) is suspended in 20 mL of acetone in a 50-mL round-bottom flask at room temperature. tert-Butyl chloroacetate 2 (4.1 mL, 28.6 mmol) is then added into the flask dropwise with stirring (approximately 5 s per drop). The reaction is stirred at room temperature for 18 h and then the potassium chloride is removed by filtration through Celite®. The solvent is then removed on a rotary evaporator (rotavap). 20 mL of diethyl ether is added to the residue, followed by washing with 5% (w/v) NaHCO3, water

333

Site-Specific Chemical Modification of a Glycoprotein Fragment Expressed in Yeast S S− K+

O

+ Cl

O

O

1

iii

S

i

O

S

3

2

O

O

NH

HN

O

S O

S

5

iv

ii

O

O

O

O

O

HS

4 O

NH

HN

OH

S O

S

6

Fig. 1. Schematic of the synthesis of the biotin thioester 6. (i) Acetone; (ii) Ethanolamine; (iii) Biotin, DMF, DIC; (iv) 95% TFA.

and brine. Next, the organic layer is separated and dried with anhydrous MgSO4. After 2 h drying, the MgSO4 is removed by filtration through Celite® and the solvent in the filtrate is removed on a rotovap to give 4.7 g of tert-butyl 2-(ethoxycarbonothioylthio)acetate 3 as a clear oil. 1H NMR (400 MHz, CDCl3): d 4.63 (q, 2 H, J = 7.2), 3.83 (s, 2H), 1.47 (s, 9H), 1.42 (t, 3H, J = 7.2). 2. tert-Butyl 2-(ethoxycarbonothioylthio)acetate 3 (4.7  g, 19.9 mmol) is placed into a 15-mL round-bottom flask with ethanolamine (1.2 g, 19.9 mmol). After 2 h stirring at room temperature, 10 mL of ethyl acetate is added into the reaction, followed by washing with 1  N HCl, water and brine. The organic layer is then separated and dried with anhydrous MgSO4. After 2 h drying, the MgSO4 is removed by filtration through Celite® and the solvent in the filtrate is removed on a rotovap. The residue is vacuum distilled, and the distillate with a bp of 35–42°C at 6 torr is collected to give 2.1 g of tert-butyl 2-mercaptoacetate 4 as a clear oil. 1H NMR (400 MHz, CDCl3): d 3.17 (d, 2 H, J = 8.0), 1.94 (t, 1 H, J = 8.0), 1.47 (s, 9H). 3. Biotin (100 mg, 0.41 mmol) is dissolved in 10 mL DMF in a 25-mL round-bottom flask. tert-Butyl 2-mercaptoacetate 4 (243  mg, 1.64  mmol) and DIC (207  mg, 1.64  mmol) are then added into the reaction. The reaction is stirred at room temperature for 24 h, and then the solvent is removed on a rotovap. The residue is purified by flash chromatography using a mixture of methanol and DCM (5:95, v/v). The resulting white solid 5 is then dissolved in 5 mL of 95% TFA and stirred at room temperature for 30 min. The solvent is then removed on a rotovap, and the residue is azeotroped three times with toluene and dried overnight under vacuum to give 115 mg of the biotin thioester 6 as a white solid (see Note 2 and ref. 6). 1H NMR (400 MHz, D2O): d 4.60 (dd, 1H, J = 7.9, 5.0), 4.42 (dd, 1H, J = 7.9, 4.6), 3.56 (s, 2 H), 3.33 (ddd, 1  H, J = 9.1, 4.6, 4.6), 2.99 (dd, 1  H, J = 12.9, 5.0), 2.77 (d, 1 H, J = 12.9), 2.68 (t, 2 H, J = 7.5), 1.75–1,39

334

Xiao and Tolbert

(m, 6  H); 13C NMR: d 205.72, 177.55, 166.92, 63.61, 61.84, 56.85, 44.47, 41.34, 35.68, 29.31, 29.19, 26.49; MALDI-FTMS: calcd for C12H18N2O4S2 (MNa+) 341.0600, found 341.0601. 3.2. Construction of the pGAP-C-IgG1 Fc Expression Plasmid

In this procedure, a DNA fragment encoding the C-IgG1 Fc ­glycoprotein is inserted into the pGAPZaA yeast expression ­plasmid, which contains a N-terminal signal peptide and the a-factor prepro leader sequence to direct protein secretion. A cysteine residue is inserted into the oligonucleotide primer complementary to the N-terminus of the protein such that it directly follows the lysine-arginine (KR) dipeptide sequence that is cleaved by the yeast Kex2 protease (see Fig. 2). 1. The pGAPzaA plasmid is double-digested with Xho I and Not I restriction enzymes at 37°C for 2 h and then treated with antarctic phosphatase at 37°C for 0.5 h. The linearized product is gel purified and stored at −20°C before usage. 2. The DNA fragment encoding the Fc region of human immunoglobulin heavy chain gamma 1 is amplified by polymerase chain reaction (PCR) using the oligonucleotide primers

a Xho I site Kex2 ggc ccg ctc gag aaa aga tgc aca ...... Gly Pro Leu Glu Lys Arg Cys Thr ......

b

α-factor prepro leader sequence

Signal peptidase +H N 3

signal peptide α-factor

Lys Arg Cys

Kex2 +H N 3

α-factor

IgG1 Fc (N-linked glycosylation site N297)

COO−

Signal peptidase cleavage in ER

Lys Arg Cys IgG1 Fc (N-linked glycosylation site N297)

COO−

Kex2 endoprotease cleavage in Golgi +H N 3

Cys IgG1 Fc (N-linked glycosylation site N297)

COO−

Fig. 2.  (a) Design of the 5¢-end of the PCR primer used to encode a cysteine residue at the N-terminal of C-IgG1 Fc adjacent to the KR dipeptide sequence that is cleaved by the yeast Kex2 protease. (b) C-IgG1 Fc expressed in yeast is fused to an N-terminal signal peptide, which directs the C-IgG1 Fc protein into the secretory pathway, and the a-factor prepro leader sequence. The a-factor leader sequence, which directs C-IgG1 Fc to be secreted, is subsequently cleaved from the fusion protein by the Kex2 protease in the Golgi apparatus to generate the N-terminal cysteine on C-IgG1 Fc.

Site-Specific Chemical Modification of a Glycoprotein Fragment Expressed in Yeast

335

5¢-GGCCCGCTCGAGAAAAGATGCACATGCCCACC GTGCCCAGCA-3¢ and 5¢-GGGCCCGCGGCGGCCGCT CATTTACCCGGAGACAGGGAGAG-3¢, with plasmid MGC-12853 as the DNA template. 3. The PCR fragment obtained from step 2 is gel purified and then double-digested with Xho I and Not I restriction enzymes at 37°C for 2 h. Once again, the resulting digestion product is gel purified. 4. The gel-purified digestion product obtained from step 3 and the linearized pGAPZaA plasmid obtained from step 1 are ligated together by T4 DNA ligase at 37°C for 2 h. 5. The ligated DNA is introduced into Top10F’ E. coli cells by electroporation, and transformed cells are selected on low-salt LB/Zeocin plates. Plasmid pGAP-C-IgG1 Fc is isolated from a single positive clone, and insertion of the correct C-IgG1 Fc-encoding DNA is confirmed by DNA sequencing. 3.3. Expression and Purification of Glycosylated C-IgG1 Fc

The glycosylated C-IgG1 Fc protein is expressed in an OCH1deleted strain of Pichia pastoris SMD1168 (7, 21). The heterogeneously glycosylated C-IgG1 Fc product is then purified using a Protein G affinity chromatography column, followed by a phenyl Sepharose® hydrophobic interaction chromatography (HIC) ­column (Fig. 3). Finally, the heterogeneously glycosylated C-IgG1 Fc is characterized by electrospray ionization (ESI) mass spectrometry (MS). 1. Plasmid pGAP-C-IgG1 Fc is linearized with restriction enzyme Avr II at 37°C for 24  h. The linearized DNA is

Fig. 3. SDS-PAGE analysis of purified C-IgG1 Fc. Lane 1: Molecular weight markers; Lane 2: C-IgG1 Fc after Protein G column purification; Lane 3: Glycosylated C-IgG1 Fc after phenyl Sepharose® column purification.

336

Xiao and Tolbert

i­ntroduced into an OCH1-deleted strain of Pichia pastoris SMD1168 by electroporation, and transformed cells are selected on YPD/Zeocin plates. The positive colonies ­containing human IgG1 Fc DNA are confirmed by PCR. 2. A single positive colony is inoculated into 2 mL of YPD/Zeocin medium and grown in a shaker at 25°C for 3 days. Following this, the 2-mL dense seed culture is inoculated into 50 mL of YPD/Zeocin medium and grown for another 3 days. 3. The 50 mL of dense culture obtained at the end of step 2 is inoculated into a spinner flask containing 1 L of supplemented YPD medium (see Note 3) with 100 mL of sterile antifoam 204. During growth of the yeast cell culture, the medium is aerated by flowing (0.2-micron) filter-sterilized air through a glass sparger at a rate of 1.5 L/min. 4. The 1-L yeast culture is incubated at 25°C for 24 h to allow the culture to reach a high cell density. Thereafter, 50 mL of a sterile 20% (w/v) glucose solution is added to the yeast culture every 24 h for 3 days. 5. The yeast cell culture is harvested on the fourth day by centrifugation at 5,400 × g for 10 min at 4°C. The recovered culture supernatant is adjusted to pH 7 with a 10  N NaOH solution, and incubated at 4°C for 2 h (during this incubation period, a precipitate often forms). Afterwards, the supernatant is filtered through filter paper using a Buchner funnel. 6. A Protein G Sepharose® column (5-mL bed volume) is preequilibrated with 50 mL of phosphate buffer, pH 7.0. 7. The filtered supernatant from step 5 is loaded onto the preequilibrated Protein G column. Following this, the column is washed with 50 mL of phosphate buffer, pH 7.0. 8. The fraction collection tubes are prepared by the addition of 0.1 mL of Tris buffer, pH 7.8 per mL of each fraction to be collected (see Note 4). 9. The C-IgG1 Fc is eluted with elution buffer and collected directly into the fraction collection tubes containing Tris buffer (prepared in step 8). The resulting fractions are analyzed by SDS-polyacrylamide gel electrophoresis (SDS-PAGE). The fractions containing C-IgG1 Fc are collected and dialyzed against Buffer A at 4°C overnight. 10. A phenyl Sepharose® column (50-mL bed volume) is preequilibrated with 100 mL of Buffer A. 11. The dialyzed C-IgG1 Fc solution obtained from step 9 is loaded onto the pre-equilibrated phenyl Sepharose® column. 12. 600 mL of a linear gradient starting with Buffer A and ending with phosphate buffer, pH 7.0 is used to elute the C-IgG1 Fc protein from the phenyl Sepharose® column. Fractions (5 mL) are collected throughout the gradient elution process and

Site-Specific Chemical Modification of a Glycoprotein Fragment Expressed in Yeast

337

analyzed by SDS-PAGE. The fractions containing only glyc­osylated C-IgG1 Fc are pooled and dialyzed against phosphate buffer, pH 7.0 at 4°C overnight (see Note 5). 13. The purified C-IgG1 Fc product is analyzed by ESI-MS. The ESI mass spectrum for heterogeneously glycosylated C-IgG1 Fc indicates the presence of high-mannose type glycoforms containing 8–12 mannose residues (see Fig. 4a). 3.4. Ligation of the Biotin Thioester to the C-IgG1 Fc

In this procedure, biotin is conjugated to the N-terminus of C-IgG1 Fc by native chemical ligation. The ligation product is then characterized by ESI-MS. 1. Approximately 1.6  mL of glycosylated C-IgG1 Fc (protein concentration » 1.5  mg/mL) is placed into a dialysis bag (6,000 MWCO) and dialyzed against hydroxylamine buffer at room temperature for 8 h (see Note 6). Following this, the protein solution is then dialyzed against ligation buffer at 4°C overnight. 2. The dialyzed protein solution is transferred into an Eppendorf tube for the native chemical ligation reaction with biotin thioester. 3. 200 mL of 40 mM biotin thioester (see Note 7) and 200 mL of freshly prepared 300  mM sodium 2-mercaptoethanesulfonate (see Note 8) are added to the Eppendorf tube containing the glycosylated C-IgG Fc protein solution. The reaction mixture is incubated at room temperature for 24 h. 4. The reaction mixture is dialyzed against phosphate buffer, pH 7.0 at 4°C for 12 h to remove the excess thioester. 5. The biotin modified, glycosylated C-IgG1 Fc is reduced by adding 5  mM DTT to the protein solution. The DTT and salts are subsequently removed by running a C18 reversedphase HPLC column with a gradient elution (10–90% acetonitrile in H2O, 0.1% TFA). 6. The purified biotin-modified, heterogeneously glycosylated C-IgG1 Fc product obtained from step 5 is analyzed by ESI-MS. The ESI mass spectrum shown in Fig. 4b indicates the formation of correctly ligated product.

3.5. Deglycosylation of Biotin-Modified Glycosylated C-IgG1 Fc

To simplify the monitoring of glycoprotein modification reactions with mass spectrometry techniques, the biotin-modified, glycosylated C-IgG1 Fc is deglycosylated by using PNGase F enzyme, and the resulting protein product is analyzed by ESI-MS. 1. 500 mL of biotin modified, glycosylated C-IgG1 Fc is placed into an Eppendorf tube. 100 units of PNGase F enzyme (0.1  mL) are added into the tube, and the reaction is incubated at room temperature for 12 h.

338

Xiao and Tolbert

Fig. 4. (a) ESI-MS analysis of heterogeneously glycosylated C-IgG1 Fc. Calcd. masses for glycoforms containing 8–12 mannose residues: 26,872, 27,035, 27,198, 27,361, 27,524, respectively; obs. 26,873, 27,034, 27,198, 27,360, 27,523, respectively. (b) ESI-MS analysis of biotin-modified, heterogeneously glycosylated C-IgG1 Fc, Calcd. masses for glycoforms containing 8–12 mannose residues: 27,098, 27,261, 27,424, 27,587, 27,750, respectively; obs. 27,098, 27,259, 27,423, 27,585, 27,748, respectively.

2. The reaction mixture is reduced with 5 mM DTT. Following this, the DTT and salts are removed by running a C18 reversed-phase HPLC column with a gradient elution ­(10–90% acetonitrile in H2O, 0.1% TFA). 3. The resulting deglycosylated protein is analyzed by ESI-MS.

Site-Specific Chemical Modification of a Glycoprotein Fragment Expressed in Yeast

339

Figure  5a shows that the removal of the N-linked ­ ligosaccharide by PNGase F greatly simplifies the mass speco trum, resulting in only a single peak for the biotin-modified, deglycosylated C-IgG1 Fc protein. The mass spectrum can also be simplified by conversion of the biotin-modified, heterogeneously glycosylated C-IgG1 Fc to a homogenous glycoform via treatment with a-mannosidase IA enzyme (see Note 9).

Fig. 5. (a) ESI-MS analysis of biotin-modified, deglycosylated C-IgG1 Fc (obtained after treatment of biotin-modified, glycosylated C-IgG1 Fc with PNGase F enzyme). Calcd. 25,395; obs. 25,392. (b) ESI-MS of biotin-modified Man5-C-IgG1 Fc (obtained after treatment of biotin-modified, glycosylated C-IgG1 Fc with a-mannosidase IA enzyme). Calcd. 26,612; obs. 26,609.

340

Xiao and Tolbert

4. Notes 1. We use tert-butyl 2-mercaptoacetate in this protocol to produce a water-soluble biotin thioester. Biotin in its protonated form is a relatively hydrophobic molecule, and biotin thioesters that do not contain an additional hydrophilic group are not water soluble. Deprotection of the tert-butyl ester in the intermediate biotin thioester produces a carboxylic acid that makes the resulting biotin thioester soluble in ligation buffer (pH 7.5). At present, tert-butyl 2-mercaptoacetate is not commercially available. However, for molecules that are initially very water soluble, a commercially available thiol (such as methyl 3-mercaptopropionate or ethyl 3-mercaptopropionate) can be used to avoid the necessity of synthesizing tert-butyl 2-mercaptoacetate. 2. Thioesters are sensitive to basic conditions and can become slowly hydrolyzed under neutral aqueous conditions. Thus, synthesized thioesters should be promptly sealed and stored in a freezer (23). 3. During the growth of the 1-L yeast culture, healthy yeast cells will produce acidic metabolites that can decrease the pH of the medium. To stabilize the pH, the medium is supplemented with 100 mM phosphate buffer, pH 6. The pH of the culture is monitored every day. If the pH falls below 6, concentrated ammonium hydroxide (28–30%, w/w) can be used to adjust the pH back up to 6. 4. The C-IgG1 Fc is collected directly into Tris buffer, pH 7.8 in order to neutralize the acidic elution buffer, and thereby prevent any potential damage to the C-IgG1 Fc protein by the acidic elution conditions. 5. A 1-L yeast culture can produce a total of approximately 30 mg of C-IgG1 Fc protein. Not all of the protein produced is completely glycosylated, and approximately 10% of the C-IgG1 Fc product is nonglycosylated. Protein G affinity purification does not separate the glycosylated and nonglycosylated forms of C-IgG Fc, whereas phenyl Sepharose® column purification can isolate the glycosylated C-IgG1 Fc. However, after the phenyl Sepharose® purification step, we typically can only recover approximately 10 mg of glycosylated C-IgG1 Fc. 6. N-terminal cysteines are very reactive chemical moieties that may react with aldehyde-containing metabolites in the yeast culture medium to form thiazolidines. To remove these thiazolidines, the protein can be treated with methoxyamine or hydroxylamine prior to performing the native chemical ligation reactions (24).

Site-Specific Chemical Modification of a Glycoprotein Fragment Expressed in Yeast

341

7. The biotin thioester is only partially soluble in water, but is completely soluble under basic pH conditions. Because of this, the 40 mM biotin thioester stock solution is freshly made in phosphate buffer, pH 7.5 right before use. 8. Sodium 2-mercaptoethanesulfonate solution is freshly prepared since the 2-mercaptoethanesulfonate may form disulfide bonds if the solution sits at room temperature for too long. 9. Although the biotin-modified C-IgG1 Fc is heterogeneously glycosylated, the glycosylation on C-IgG1 Fc can be further manipulated to homogeneous glycoforms by using an in vitro enzymatic process. One example of the conversion of biotinmodified, heterogeneously glycosylated C-IgG1 Fc to homogeneously glycosylated biotin-Man5-C-IgG1 Fc by a-mannosidase IA treatment is shown in Fig. 5b.

Acknowledgment This work was supported by Award Number R01GM090080 from the National Institute of General Medical Sciences. References 1. Hackenberger, C. P. R., and Schwarzer, D. (2008) Chemoselective ligation and modification strategies for peptides and proteins. Angew. Chem. Int. Ed. 47, 10030–10074. 2. Flavell, R. R., and Muir, T. W. (2009) Expressed protein ligation (EPL) in the study of signal transduction, ion conduction, and chromatin biology. Acc. Chem. Res. 42, 107–116. 3. Xiao, J., Burn, A., and Tolbert, T. J. (2008) Increasing Solubility of Proteins and Peptides by Site-Specific Modification with Betaine. Bioconjugate Chem. 19, 1113–1118. 4. Kochendoerfer, G. G. (2005) Site-specific polymer modification of therapeutic proteins. Curr. Opin. Chem. Biol. 9, 555–560. 5. Tolbert, T. J., and Wong, C.-H. (2000) InteinMediated Synthesis of Proteins Containing Carbohydrates and Other Molecular Probes. J. Am. Chem. Soc. 122, 5421–5428. 6. Tolbert, T. J., and Wong, C.-H. (2002) New methods for proteomic research: preparation of proteins with N-terminal cysteines for labeling and conjugation. Angew. Chem. Int. Ed. 41, 2171–2174. 7. Xiao, J., Chen, R., Pawlicki, M. A., and Tolbert, T. J. (2009) Targeting a homogeneously glycosylated antibody Fc to bind cancer cells

using a synthetic receptor ligand. J. Am. Chem. Soc. 131, 13616–13618. 8. Dawson, P. E., Muir, T. W., Clark-Lewis, I., and Kent, S. B. (1994) Synthesis of proteins by native chemical ligation. Science 266, 776–779. 9. Erlanson, D. A., Chytil, M., and Verdine, G. L. (1996) The leucine zipper domain controls the orientation of AP-1 in the NFAT.AP-1.DNA complex. Chem. Biol. 3, 981–991. 10. Tolbert, T. J., Franke, D., and Wong, C. H. (2005) A new strategy for glycoprotein synthesis: ligation of synthetic glycopeptides with truncated proteins expressed in E. coli as TEV protease cleavable fusion protein. Bioorg. Med. Chem. 13, 909–915. 11. Marsac, Y., Cramer, J., Olschewski, D., Alexandrov, K., and Becker, C. F. (2006) Sitespecific attachment of polyethylene glycol-like oligomers to proteins and peptides. Bioconjugate Chem. 17, 1492–1498. 12. Busch, G. K., Tate, E. W., Gaffney, P. R., Rosivatz, E., Woscholski, R., and Leatherbarrow, R. J. (2008) Specific N-terminal protein ­labelling: use of FMDV 3C pro protease and native chemical ligation. Chem. Commun., 3369–3371.

342

Xiao and Tolbert

13. Xiao, J., and Tolbert, T. J. (2009) Synthesis of Polymerizable Protein Monomers for ProteinAcrylamide Hydrogel Formation. Biomacromolecules 10, 1939–1946. 14. Xiao, J., and Tolbert, T. J. (2009) Synthesis of N-terminally linked protein dimers and trimers by a combined native chemical ligationCuAAC click chemistry strategy. Org Lett 11, 4144–4147. 15. Liu, C. F., and Tam, J. P. (1994) Peptide segment ligation strategy without use of protecting groups. Proc. Natl. Acad. Sci. U.S.A. 91, 6584–6588. 16. Tam, J. P., and Yu, Q. (2002) A facile ligation approach to prepare three-helix bundles of HIV fusion-state protein mimetics. Org. Lett. 4, 4167–4170. 17. Ren, H., Xiao, F., Zhan, K., Kim, Y. P., Xie, H., Xia, Z., and Rao, J. (2009) A Biocompatible Condensation Reaction for the Labeling of Terminal Cysteine Residues on Proteins. Angew. Chem., Int. Ed. 48, 9658–9662. 18. Dunn, M. J. (1994) Detection of proteins on blots using the avidin-biotin system. Methods. Mol. Biol. 32, 227–232.

19. Lesaicherre, M. L., Lue, R. Y., Chen, G. Y., Zhu, Q., and Yao, S. Q. (2002) Inteinmediated biotinylation of proteins and its application in a protein microarray. J. Am. Chem. Soc. 124, 8768–8769. 20. Strausberg, R. L., Feingold, E. A., Klausner, R. D., and Collins, F. S. (1999) The mammalian gene collection. Science 286, 455–457. 21. Cregg, J. M., Tolstorukov, I., Kusari, A., Sunga, J., Madden, K., and Chappell, T. (2009) Expression in the yeast Pichia pastoris. Methods Enzymol. 463, 169–189. 22. Woulfe, S. R., and Miller, M. J. (1986) The synthesis of substituted ((3(S)-(acylamino)2-oxo-1-azetidinyl)thio)acetic acids. J. Org. Chem. 51, 3133–3139. 23. Tolbert, T. J., and Wong, C. H. (2004) Conjugation of glycopeptide thioesters to expressed protein fragments: semisynthesis of glycosylated interleukin-2. Methods Mol. Biol. 283, 255–266. 24. Villain, M., Vizzavona, J., and Rose, K. (2001) Covalent capture: a new tool for the purification of synthetic and recombinant polypeptides. Chem. Biol. 8, 673–679.

Chapter 22 On-Resin Convergent Synthesis of a Glycopeptide from HIV gp120 Containing a High Mannose Type N-Linked Oligosaccharide Rui Chen and Thomas J. Tolbert Abstract This chapter describes a rapid and efficient approach for the solid-phase synthesis of N-linked glycopeptides that utilizes on-resin glycosylamine coupling to produce N-linked glycosylation sites. In this method, the full-length nonglycosylated peptide is first synthesized on a solid-phase support using standard Fmoc chemistry. The glycosylation site is then introduced through an orthogonally protected 2-phenylisopropyl (PhiPr) aspartic acid (Asp) residue. After selective deprotection of the Asp residue, a high mannose type oligosaccharide glycosylamine is coupled on-resin to the free Asp side chain to form a N-glycosidic bond. Subsequent protecting group removal and peptide cleavage from the resin ultimately yields the desired glycopeptide. This strategy provides an effective route for conducting glycosylation reactions on a solid-phase support, simplifies the process of glycopeptide purification relative to solution-phase glycopeptide synthesis strategies, and enables the recovery of potentially valuable, unreacted oligosaccharides. This approach has been applied to the solid-phase synthesis of the N-linked high mannose glycosylated form of peptide T (ASTTTNYT), a fragment of the HIV-1 envelope glycoprotein gp120. Key words: N-Linked glycopeptide, Solid-phase peptide synthesis, High mannose oligosaccharide, 2-Phenylisopropyl protecting group, Glycosylamine, Peptide T, HIV, gp120

1. Introduction Homogenous N-linked glycopeptides are valuable materials for biochemical, structural and medicinal studies to help understand the specific roles of N-linked glycosylation (1) and facilitate the development of glycopeptides-based vaccines and therapeutics (2). However, the starting materials for glycopeptide synthesis, large N-linked oligosaccharides, are often difficult to obtain from

Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_22, © Springer Science+Business Media, LLC 2011

343

344

Chen and Tolbert

biological systems and are also difficult to chemically synthesize (3–8). In addition, the coupling of large, bulky N-linked oligosaccharides to peptides can be a low-yielding reaction that is prone to aspartimide formation (9, 10). Because of this, it is crucial to utilize an efficient method for N-linked glycopeptide synthesis that does not waste valuable N-linked oligosaccharide starting materials. The method of N-linked glycopeptide synthesis (11) described in this chapter effectively combines the advantages of both solidphase peptide synthesis (SPPS) (12) and in-solution Lansbury aspartylation (13, 14) by introducing the glycosylation site via glycosylamine coupling to partially protected, full-length peptides on a solid-phase support (11). As illustrated in Fig. 1, the full-length nonglycosylated peptide sequence is first built via SPPS. At the desired glycosylation site, a 2-phenylisopropyl (PhiPr) (15) orthogonally protected Asp residue is incorporated. The PhiPr protecting group efficiently suppresses aspartimide formation during the construction of the peptide, unlike other types of orthogonal protecting groups such as allyl esters (11, 16–18). Next, the PhiPr protecting group is selectively removed on-resin to free the carboxylic acid side chain of the Asp residue, which allows the introduction of the N-linked oligosaccharide. Then a glycosylamine, such as Man8GlcNAc2NH2, 2, is coupled to the free Asp side chain on-resin. This key step has been optimized for both the monosaccharide N-acetylglucosamine (GlcNAcNH2) and a high mannose oligosaccharide (Man8GlcNAc2NH2, 2) on aspartimide prone peptide sequences, and satisfactory glycosylation yields have been achieved (11). At the end of the synthesis procedure, final deprotection and resin cleavage provide the desired N-linked glycopeptide product. An important aspect of this strategy is that excess N-linked oligosaccharides can be utilized to drive the glycosylation reaction on-resin, and un-reacted, valuable oligosaccharide can be conveniently recovered after the on-resin glycosylamine

Fig. 1. Strategy for on-resin convergent synthesis of N-linked high mannose oligosaccharide-containing glycopeptides.

On-Resin Convergent Synthesis of a Glycopeptide

345

coupling reaction by filtration and silica gel purification. The recovered oligosaccharide can then be reused in later syntheses, increasing the overall efficiency of this method (11). Here, this strategy for N-linked glycopeptide synthesis is demonstrated by the synthesis of a glycosylated form of peptide T (ASTTTNYT) containing the high mannose oligosaccharide Man8GlcNAc2, 1. Peptide T is a fragment of the HIV-1 envelope protein gp120 (consisting of residues 185–192) that contains a naturally occurring N-linked glycosylation site consensus sequence (19, 20).

2. Materials 1. Milli-Q water (purified and deionized by a Millipore water purification system). 2. Saturated, aqueous ammonium hydrogen carbonate solution (produced with Milli-Q water). 3. Standard amino acid building blocks for peptide synthesis: The amino acids side chains are protected by the most commonly used protecting groups, including tert-butyl (tBu) for Tyr, Ser, Thr, Glu and Asp; 2,2,4,6,7-pentamethyldihydrobenzofuran-5-sulfonyl (Pbf) for Arg; trityl (trt) for Gln, Asn and Cys, His; and tert-butoxycarbonyl (Boc) for Trp (AAPPTEC, Louisville, KY). 4. Fmoc-Asp(O-2PhiPr)-OH (EMD Biosciences, Gibbstown, NJ). 5. Man8GlcNAc2, produced in yeast (see Note 1). 6. 3-(Diethoxyphosphoryloxy)-1, 2, 3-benzotriazin-4(3 H)-one (DEPBT) (AAPPTEC). 7. 1-Hydroxybenzotriazole (HOBt). 8. N,N ´-Diisopropylcarbodiimide (DIC). 9. N,N-Diisopropylethylamine (DIEA). 10. Acetic anhydride. 11. N-methylmorpholine (NMM). 12. Piperidine. 13. Trifluoroacetic acid (TFA). 14. Dichloromethane (DCM). 15. Dimethylformamide (DMF). 16. Methyl sulfoxide. 17. Acetonitrile (ACN). 18. 0.1% (v/v) TFA in acetonitrile, for HPLC and spectro­ photometry.

346

Chen and Tolbert

19. Triisopropylsilane (TIS) (Sigma-Aldrich, St. Louis, MO). 20. Diethyl ether anhydrous (ether absolute). 21. Methanol. 22. Ethyl acetate. 23. PEG-polyacrylamide (PEGA)-based Rink Amide resin (50–100 mesh, 0.20–0.50 mmol/g) (EMD Biosciences). 24. Poly-Prep® chromatography columns (0.8 × 4 cm) (Bio-Rad, Hercules, CA). 25. Economy Mini-Spin Columns (0.8  mL resin capacity) (Thermo Scientific, Waltham, MA). 26. Silica gel 62, 60–200 mesh (EMD). 27. Silica gel TLC plates, Si 60F254 (EMD Chemicals). 28. Ninhydrin test solution A: Dissolve 40 g of phenol in 20 ml of ethanol; Ninhydrin test solution B: Add 1 ml of KCN buffer (dissolve 16.5 mg of KCN in 25 ml of water) into 49 ml of pyridine; Ninhydrin test solution C: Dissolve 1.0 g of ninhydrin in 20 ml of ethanol (21). 29. Vanillin staining solution: Dissolve 15 g of vanillin in 250 mL of ethanol and 2.5 mL of concentrated sulfuric acid. 30. Phenol-sulfuric acid test solution: Mix together 50  mL of concentrated sulfuric acid, 20  mL of sample and 1  mL of phenol (22). 31. Peptide synthesizer (Applied Biosystems 433A Peptide Synthesizer). 32. Semi-preparative reversed-phase high-performance liquid chromatography (RP-HPLC) system equipped with a Hyper Prep 120C18 column (8 mm, Length 250 mm, I.D. 10 mm) (Alltech Associates, Inc.). 33. Lyophilizer (VirTis Sentry 2.0). 34. UV-visible spectrophotometer. 35. Nuclear magnetic resonance (NMR) spectrometer. 36. Matrix-assisted laser desorption/ionization-time of flight (MALDI-TOF) mass spectrometer.

3. Methods The following method, based upon the work recently reported by Chen and Tolbert (11), describes the on-resin convergent synthesis of high mannose containing N-linked glycopeptides. The individual procedures consist of (1) preparation of glycosylamines from free reducing end sugars; (2) solid-phase synthesis of a

On-Resin Convergent Synthesis of a Glycopeptide

347

f­ull-length nonglycosylated peptide, in which an Asp residue is protected orthogonally by the PhiPr group at the desired glycosylation site; (3) selective deprotection of the Asp side chain; (4) on-resin coupling of the glycosylamine; (5) final deprotection, resin cleavage and purification; and (6) recovery of un-reacted high mannose oligosaccharide. 3.1. Preparation of Glycosylamines

The conversion of free reducing end sugars into their corresponding glycosylamines is accomplished by treatment with saturated, aqueous ammonium hydrogen carbonate at room temperature (23). The residual ammonium hydrogen carbonate is removed by repetitive rotary evaporation and lyophilization to give the desired glycosylamine for direct use in glycosylation reactions without further purification. The percentage of glycosylamine conversion can be monitored by 1H-NMR during this process. 1. The free reducing end sugar, Man8GlcNAc2, 1 (20–100 mg), was dissolved in 25 mL of a saturated, aqueous ammonium hydrogen carbonate solution and stirred for 5–6 days at room temperature. 2. 1H-NMR was utilized to monitor the reaction, and showed that the conversion of the free sugar into its corresponding b-dglycosylamine occurred with a very high yield (see Fig. 2). 3. The reaction solution was diluted with Milli-Q water and rotary evaporated at room temperature to dryness. This step was repeated multiple times to remove the residual ammonium hydrogen carbonate. 4. The sample was dissolved in Milli-Q water again and lyophilized repeatedly until the sample reached a constant mass. 5. Glycosylamines prepared in this manner can be stored at −20°C in a desiccator for later use.

Fig. 2. Expanded 400 MHz 1H-NMR spectra (in D2O) of (a) free reducing end Man8GlcNAc2 1, and (b) the resulting glycosylamine Man8GlcNAc2NH2 2 produced after 5  days of treatment of 1 with saturated aqueous ammonium hydrogen carbonate.

348

Chen and Tolbert

3.2. Solid-Phase Synthesis of Full-Length Nonglycosylated Peptides

The synthesis of full-length, protected nonglycosylated peptides is carried out on a peptide synthesizer with standard Fmoc methods utilizing DIC/HOBt coupling on Rink Amide PEGA resin. Manual coupling and capping is preferred for the loading of the first amino acid residue. At the desired glycosylation site, FmocAsp(O-2PhiPr)-OH is added in the sequence to allow for selective deprotection after the full-length peptide is synthesized and the introduction of glycosylation. 1. Rink Amide PEGA resin (0.001–0.01  mmol) (see Note 2) was washed with DMF and then preswelled in DMF in a Poly-Prep® chromatography column, which was utilized as the reaction vessel. After 0.5 h, the DMF was drained off. 2. The first Fmoc-amino acid (5 eq) and 5 eq of DEPBT were dissolved in a minimum volume of DMF, and then added into the reaction vessel. Next, 3 eq of DIEA were added. 3. The reaction vessel was capped and attached to a rotator. The rotator was rotated at 30  rpm at room temperature for approximately 3  h. A ninhydrin test (see Note 3) was then performed to determine the completion of the amino acid loading. The resin was washed consecutively with DMF (two times), dry DCM (three times) and then with DMF again (two times). 4. Treatment with a large excess of acetic anhydride (1.5  ml) with 3 eq of N-methylmorpholine in DCM for 45 min was then used to cap the unreacted amino groups. 5. The resulting resin was then washed consecutively with DMF (two times), dry DCM (three times) and then with DMF again (two times). The Fmoc protecting groups were removed using 20% (w/v) piperidine in DMF (10 min × 3). The loading of the first amino acid residue was determined by monitoring the removal of the Fmoc protecting group using a UV-visible spectrophotometer (l = 301 nm; e = 7,800 cm−1 M−1) (see Note 4) (24). 6. The resin was then transferred into the reaction vessel of an automated peptide synthesizer. The subsequent synthesis of the full-length peptide was carried out on the peptide synthesizer by standard Fmoc methods utilizing DIC/HOBt. At the glycosylation site, Fmoc-Asp(O-2PhiPr)-OH was incorporated. All the other amino acids side chains were protected by groups that are commonly employed in Fmoc SPPS protocols. The N-terminal Fmoc was kept on the peptide at the end of the synthesis procedure. 7. The resin was washed consecutively with DMF (two times), dry DCM (three times), DMF (two times) and finally with dry DCM, and transferred into a Poly-Prep® chromatography column for the next reaction step. (Alternatively, the resin can also be stored at 4°C for later use.)

On-Resin Convergent Synthesis of a Glycopeptide

3.3. Selective Deprotection of the Asp Side Chain

349

The PhiPr protected Asp residue is selectively deprotected by treatment with 1% (v/v) TFA in DCM. Under these conditions the other amino acid side chains remain protected, thereby selectively creating a free carboxylic acid at the site where ­glycosylamine coupling will be utilized to create a N-linked glycosylation site (see Note 5). 1. The resin from Subheading 3.2 was swelled with DCM in the reaction column for 10 min. Excess DCM was then removed. 2. The reaction column was filled with a solution of DCM/ TFA/TIS (94:1:5, v/v/v), capped and shaken for 3 min. The reaction solution was then drained off. This step was repeated three times. 3. The resin was then washed thoroughly with DCM (five times), DMF (five times) and DMSO (five times). 4. The selectively deprotected resin was transferred into an Economy Mini-Spin Column before proceeding to the next reaction (described below in Subheading 3.4).

3.4. On-Resin Coupling of the Glycosylamine

On-resin glycosylamine coupling is the key step of this glycopeptide synthesis method. The prepared glycosylamine, in this case glycosylamine 2, is coupled to the free Asp acid side chain, which was selectively deprotected in Subheading 3.3, to form a N-linked glycosidic bond on the resin-bound peptide. This step is carried out in a Mini-Spin column to minimize the reaction volume. The reaction conditions, including the coupling reagents, solvents, resins, amounts of base and equivalents of glycosylamines were optimized as described previously (11). 1. The selectively deprotected resin from Subheading  3.3 was swelled with DMSO in a Mini-Spin reaction column for 10 min. Excess DMSO was then removed. 2. 3 eq of glycosylamine 2 and 3 eq of DEPBT were dissolved in a minimum amount of DMSO (100–200  mL) in separate Eppendorf tubes, and then added into the mini reaction column. A minimum amount of DMSO was then used to rinse the residual glycosylamine left in the Eppendorf tube, and this was also added into the reaction column. The total reaction volume was kept at less than one third of the capacity of the Mini-Spin column (see Note 6). 3. The reaction column was attached to a rotator and the coupling reaction was rotated at 30 rpm at room temperature for 12 h (see Note 7). 4. The reaction solution was collected by centrifugation and filtration. The resin was then washed with DMSO (two times), and this wash solution was combined with the reaction solution. The combined solutions can be utilized to recover unreacted high mannose oligosaccharide as described in Subheading 3.6.

350

Chen and Tolbert

Fig. 3. Analytical HPLC chromatogram profile for crude peptides after on-resin ­coupling of glycosylamine Man8GlcNAc2NH2 2 to the peptide Fmoc-ASTTTNYT, showing a 65% glycosylation yield.

Following this, the resin was washed further with DMF (three times), dry DCM (three times), and DMF (two times). 5. The resin was transferred into a Poly-Prep® chromatography column for final deprotection and resin cleavage as described in Subheading 3.5 below. 6. For the purpose of determining the glycosylation yields, the crude peptides without the N-terminal Fmoc protecting groups removed were cleaved from a resin sample, precipitated, washed twice with anhydrous diethyl ether, and then analyzed by HPLC. The glycosylation yields were determined by quantification of the analytical HPLC signals at l = 254 nm (Fig. 3; see Note 8). 3.5. Final Deprotection, Resin Cleavage, and Product Purification

The final glycopeptide product is obtained by the following procedure: The N-terminal Fmoc is first removed using 20% (v/v) piperidine in DMF. The glycopeptide is then fully deprotected and cleaved off the resin by treatment with neat TFA containing 1% (v/v) TIS scavenger. Crude glycopeptide after resin cleavage is purified by diethyl ether precipitation and then by preparative RP-HPLC. 1. The resin from Subheading 3.4 was swelled with DMF for 10 min in a Poly-Prep® column. The DMF was then drained off. 2. The reaction column was filled with a solution of 20% (v/v) piperidine in DMF, capped and rotated at 30 rpm for 10 min. The resin was then washed with DMF. This step was repeated twice, and then a Ninhydrin test was conducted to confirm the removal of the Fmoc protecting group. 3. The resin was then washed thoroughly with DMF (three times), dry DCM (two times), DMF (three times) and dry DCM (two times). 4. The reaction column was next filled with a solution of TFA containing 1% (v/v) TIS scavenger and sealed. This step was carried out under a static incubation mode (i.e., without rotation) for 2–3 h.

On-Resin Convergent Synthesis of a Glycopeptide

351

Fig. 4. MALDI-TOF MS spectrum of the glycopeptide Man8GlcNAc2-Peptide T. Expected mass: [M + Na]+ = 2582.97.

5. The TFA solution was collected by filtration and rotary evaporated to a volume of about 2  ml. An excess of anhydrous diethyl ether was then added to this solution to precipitate the peptides that were cleaved from the resin. The ether solution was cooled to −20°C for 0.5  h to allow for maximum precipitation of the cleaved peptides. 6. The ether solution was centrifuged at 10,000 × g in a 50-mL centrifuge tube. The supernatant was discarded. The collected precipitate was resuspended in anhydrous diethyl ether and centrifuged again. This step was repeated, and then the washed precipitate was air dried in a fume hood for 5 min. 7. For purification by semi-preparative HPLC, the precipitated peptides were redissolved in a loading buffer containing 20% (v/v) ACN and 1% (v/v) acetic acid in water. After being loaded onto a RP-HPLC C18 column, the peptides were purified by elution with an acetonitrile gradient containing 0.1% (v/v) TFA (see Note 9). 8. The preparative HPLC fractions were analyzed by analytical HPLC and MALDI-TOF mass spectrometry (MS) to confirm their purity and peptide masses (see Fig. 4). 9. The purified glycopeptides were lyophilized and stored at −20°C. 3.6. Recovery of Unreacted High Mannose Oligosaccharide

One of the important benefits of the on-resin glycosylamine coupling approach described herein is the possibility of conveniently recovering valuable, unreacted oligosaccharides after performing the glycosylation reaction (described in Subheading 3.4). By collecting

352

Chen and Tolbert

the glycosylation reaction solution by filtration, and then performing a precipitation and silica gel purification step, some of the valuable N-linked oligosaccharides can be recovered. 1. The pooled reaction solutions (from Subheading 3.4, step 4 of the on-resin glycosylamine coupling procedure) containing unused oligosaccharide in DMSO were precipitated with anhydrous diethyl ether. 2. The precipitate was then spun down by centrifugation and washed twice with diethyl ether. 3. The precipitate was then redissolved in a minimum amount of an ethyl acetate/methanol/water (10:3:3, v/v/v) solution, and then loaded onto an appropriately sized flash silica column (see Note 10). 4. The silica column was washed with an excess of ethyl acetate/ methanol/water solution (10:3:3, v/v/v). Next, a gentle gradient of ethyl acetate/methanol/water (10:3:3 to 5:3:3, v/v/v) was introduced and the eluted fractions were collected from the column. 5. A phenol-sulfuric acid test was used to detect the glycan-containing fractions (see Note 11) (22). The positive fractions were pooled together and lyophilized to give Man8GlcNAc2 1, with an approximate yield of 78% (based on the number of extra equivalents of Man8GlcNAc2NH2 2 used in the glycosylation reaction).

4. Notes 1. The human-type Man8GlcNAc2 high mannose oligosaccharide was produced in our laboratory using a glycosylationdeficient yeast strain (manuscript in preparation). 2. PEGA resins are sold in wet form, preswelled in methanol. The amount (mmol) of resin needed was determined by the wet weight of the PEGA resin, using the manufacturer’s reported substitution to convert weight to millimoles of substituted resin. 3. Ninhydrin test (21): First, the resin is washed consecutively with DMF (three times) and dry DCM (three times). A small amount of resin (an aliquot 20 mL of resin in DCM) is then transferred into a glass test tube. Next, 30  mL of each Ninhydrin test solution (A, B, and C; see Subheading 2) are added. The glass test tube is then placed in a heating block at 95°C for at least 5 min. The appearance of a bluish color indicates the presence of unreacted free amines, and shows that the reactions are incomplete.

On-Resin Convergent Synthesis of a Glycopeptide

353

4. In order to quantify the Fmoc removal, the 20% (v/v) piperidine (in DMF) solutions need to be collected and pooled together. The loading can be determined by the following equation: Loading (mmol/g) = (A/el ) × V/m, where A is the absorbance at 301 nm of the Fmoc removal solution (blanked by using 20% (v/v) piperidine in DMF); e = 7,800 cm−1 M−1; l = 1 cm, the path length of the cuvette; V is the total volume (mL) of Fmoc removal solution; and m is the mass (g) of the resin (24). 5. It is preferable to remove the PhiPr group immediately before on-resin glycosylamine coupling (Subheading 3.4). 6. The Mini-Spin columns need to be sealed carefully to prevent leaking of the reaction solutions and loss of valuable reagents. 7. By shortening the coupling time to 4  h, we have observed slightly lower glycosylation yields. Conversely, by extending the coupling reaction time to 24 h, we have observed slightly better glycosylation yields. 8. The peptide’s N-terminal Fmoc was left on in order to aid quantification by analytical HPLC using UV detection at l = 254 nm. The resulting crude, precipitated peptides were then analyzed by analytical HPLC (0–80% Buffer B; 10 min; Beckman SGB 0.46 × 5  cm Zorbax C8 column; Buffer A: 0.1% (v/v) TFA in water; Buffer B: 90% (v/v) acetonitrile /10% (v/v) water/0.1% (v/v) TFA; detection at 254  nm). Individual peaks were identified by mass spectrometry and the distribution of products (glycopeptide:uncoupled peptide:aspartimide and its adduct) was determined by integration of the analytical HPLC absorption signals using Peaksimple 2000 version 2.83 software (SRI Instruments). The glycosylation yields were calculated as the percentage of glycopeptide in the crude product. 9. The acetonitrile gradient varies depending on the particular glycopeptides being purified. It is recommended that an analytical HPLC trial run be carried out prior to preparative HPLC purification to guide the development of the final purification procedure. 10. The amount of silica gel used is 10- to 100-fold more than the crude sample by weight. The gradient needs to be gentle to allow effective separation of the product oligosaccharide from impurities. Usually, Man8GlcNAc2 did not elute until the gradient polarity was increased to 5:3:3 (v/v/v) ethyl acetate/methanol/water. 11. Phenol-sulfuric acid test (22): 20  mL sample, 1  mL phenol, and 50 mL concentrated sulfuric acid were added sequentially and mixed thoroughly. A yellow-orange color indicates the presence of glycan. If the color changes are very subtle, incubation at 37°C can help to facilitate color development.

354

Chen and Tolbert

Acknowledgment This work was supported by Award Number R01GM090080 from the National Institute of General Medical Sciences. References 1. Dove, A. (2001) The bittersweet promise of glycobiology. Nature Biotech. 19, 913–917. 2. Scanlan, C. N., Offer, J., Zitzmann, N., and Dwek, R. A. (2007) Exploiting the defensive sugars of HIV-1 for drug and vaccine design. Nature 446, 1038–1045. 3. Lee, J. C., Greenberg, W. A., and Wong, C. H. (2006) Programmable reactivity-based onepot oligosaccharide synthesis. Nat. Protoc. 1, 3143–3152. 4. Lis, H., and Sharon, N. (1978) Soybean agglutinin--a plant glycoprotein. Structure of the carboxydrate unit. J. Biol. Chem. 253, 3468–3476. 5. Dudkin, V. Y., Miller, J. S., and Danishefsky, S. J. (2004) Chemical synthesis of normal and transformed PSA glycopeptides. J. Am. Chem. Soc. 126, 736–738. 6. Demchenko, A. V. (2005) Strategic approach to the chemical synthesis of oligosaccharides. Lett. Org. Chem. 2, 580–589. 7. Chen, R., Pawlicki, M. A., Hamilton, B. S., and Tolbert, T. J. (2008) Enzyme-catalyzed synthesis of a hybrid N-linked oligosaccharide using N-acetylglucosaminyltransferase I. Adv. Synth. Catal. 350, 1689–1695. 8. Seeberger, P. H. (2008) Automated oligosaccharide synthesis. Chem. Soc. Rev. 37, 19–28. 9. Kan, C., Trzupek, J. D., Wu, B., Wan, Q., Chen, G., Tan, Z., Yuan, Y., and Danishefsky, S. J. (2009) Toward homogeneous erythropoietin: Chemical synthesis of the Ala(1)-Gly(28) glycopeptide domain by “alanine” ligation. J. Am. Chem. Soc. 131, 5438–5443. 10. Tan, Z., Shang, S., Halkina, T., Yuan, Y., and  Danishefsky, S. J. (2009) Toward Homogeneous Erythropoietin: Non-NCLBased Chemical Synthesis of the Gln(78)Arg(166) Glycopeptide Domain. J. Am. Chem. Soc. 131, 5424–5431. 11. Chen, R., and Tolbert, T. J. (2010) Study of On-resin Convergent Synthesis of N-linked Glycopeptides Containing a Large High Mannose Oligosaccharide. J. Am. Chem. Soc. 132, 3211–3216. 12. Hojo, H., and Nakahara, Y. (2007) Recent progress in the field of glycopeptide synthesis. Biopolymers 88, 308–324.

13. Cohen-Anisfeld, S. T., and Lansbury, P. T. (1993) A practical, convergent method for glycopeptide synthesis. J. Am. Chem. Soc. 115, 10531–10537. 14. Mandal, M., Dudkin, V. Y., Geng, X. D., and Danishefsky, S. (2004) In pursuit of ­carbohydrate-based HIV vaccines, Part 1: The total synthesis of hybrid-type gp120 fragments. Angew. Chem. Int. Ed. Engl. 43, 2557–2561. 15. Yue, C. W., Thierry, J., and Potier, P. (1993) 2-Phenyl isopropyl esters as carboxyl terminus protecting groups in the fast synthesis of peptide-fragments. Tetrahedron Lett. 34, 323–326. 16. Kates, S. A., Delatorre, B. G., Eritja, R., and Albericio, F. (1994) Solid-phase N-glycopeptide synthesis using allyl side-chain protected Fmoc-amino acids. Tetrahedron Lett. 35, 1033–1034. 17. Offer, J., Quibell, M., and Johnson, T. (1996) On-resin solid-phase synthesis of asparagine N-linked glycopeptides: Use of N-(2-acetoxy4-methoxybenzyl) (AcHmb) aspartyl amidebond protection to prevent unwanted aspartimide formation. J. Chem. Soc., Perkin Trans. 1, 175–182. 18. Packman, L. C. (1995) N-2 Hydroxy-4Methoxybenzyl (Hmb) Backbone protection strategy prevents double aspartimide formation in a difficult peptide sequence. Tetrahedron Lett. 36, 7523–7526. 19. Pert, C. B., Hill, J. M., Ruff, M. R., Berman, R. M., Robey, W. G., Arthur, L. O., Ruscetti, F. W., and Farrar, W. L. (1986) Octapeptides deduced from the neuropeptide receptor-like pattern of antigen T4 in brain potently inhibit human immunodeficiency virus receptor binding and T-cell infectivity. Proc. Natl. Acad. Sci. USA 83, 9254–9258. 20. D’Ursi, A., Caliendo, G., Perissutti, E., Santagada, V., Severino, B., Albrizio, S., Bifulco, G., Spisani, S., and Temussi, P. A. (2007) Conformation-activity relationship of peptide T and new pseudocyclic hexapeptide analogs. J. Pept. Sci. 13, 413–421. 21. Kaiser, E., Colescot, RL, Bossinge,CD, and Cook, P. I. (1970) Color test for detection of free terminal amino groups in solid-phase

On-Resin Convergent Synthesis of a Glycopeptide s­ ynthesis of peptides. Anal. Biochem. 34, 595–598. 22. Dubois, M., Gilles, K. A., Hamilton, J. K., Rebers, P. A., and Smith, F. (1956) Colorimetric method for determination of sugars and related substances. Anal. Chem. 28, 350–356. 23. Likhosherstov, L. M., Novikova, O. S., Derevitskaja, V. A., and Kochetkov, N. K.

355

(1986) A new simple synthesis of amino sugar beta-d-glycosylamines, Carbohydr. Res. 146, C1–C5. 24. Kay, C., Lorthioir, O. E., Parr, N. J., Congreve, M., McKeown, S. C., Scicinski, J. J., and Ley, S. V. (2000) Solid-phase reaction monitoring – Chemical derivatization and off-bead analysis, Biotechnol. Bioeng. 71, 110–118.

Chapter 23 Design and Synthesis of Novel Functional Lipid-Based Bioconjugates for Drug Delivery and Other Applications Rupa R. Sawant and Vladimir P. Torchilin Abstract The modification of biologicals such as proteins/peptides, small molecules, and other polymers with lipids provides an efficient method for mediating their insertion into liposomes and lipid-core micellar nanocarriers. In this chapter, we describe several representative protocols developed in our laboratory for the bioconjugation of liposomes and lipid-core micelles for drug/gene delivery and diagnostic imaging applications. Key words: Micelles, Liposomes, Immunomicelles, Antibody, Peptide, Ascorbic acid, Polyethyleneimine, DNA, Polychelating amphiphilic polymer

1. Introduction The conjugation of lipids with peptides/proteins, small molecules, and other polymers is an efficient way to generate biologicals that can be efficiently inserted into drug delivery systems. The most popular and well-investigated drug delivery systems are based on liposomes (mainly for water-soluble drugs) and lipid-core polymeric micelles (mainly for drugs that are poorly soluble in water) (Fig. 1). Liposomes are artificial phospholipid vesicles that can be loaded with a variety of drugs. Their size usually varies from 50 to 1,000 nm (1, 2). Liposomes can be made long-circulating (i.e., have the ability to stay in the blood for a prolonged period), by the addition of a polyethylene glycol (PEG) coating (3), to allow their accumulation in areas of the body with a compromised (leaky) vasculature, such as tumors and infarcts, by virtue of the enhanced permeability and retention (EPR) effect.

Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_23, © Springer Science+Business Media, LLC 2011

357

358

Sawant and Torchilin Poorly soluble drug Water soluble drug

Liposome

Micelle

Fig. 1. Schematic representation of the interaction of drug molecules with liposomes and micelles.

Micelles – particularly those comprised of PEG-phosphati­ dylethanolamine (PEG-PE) – are another type of nanocarrier structure of special interest. Here, the use of lipid moieties as hydrophobic blocks capping hydrophilic polymer chains (e.g., PEG) can provide additional advantages for particle stability when compared with conventional amphiphilic polymer micelles due to the presence of two fatty acid acyl groups, which contribute considerably to an increase in the hydrophobic interactions between the polymeric chains in the micelle core (4). These types of micelles are structured in such a way that the outer PEG corona, known to be highly water-soluble and highly hydrated, serves as an effective steric protector in biological media. In contrast, the phospholipid residues, which represent the micelle core, are extremely hydrophobic and can solubilize drugs that are poorly soluble in aqueous media (e.g., paclitaxel (5), camptothecin (6), porphyrins (7), vitamin K3 (8), and others). In general, PEG-PE micelles demonstrate good stability, longevity, and an ability to efficiently accumulate in areas with a damaged vasculature via the EPR effect (4, 9, 10). For use in biomedical applications, the stable attachment of ligands (e.g., proteins, peptides, or small molecules) is an important element that makes liposome- and micelle-based nanocarriers targetable. Typically, these ligands are attached to the water-exposed PEG terminus to allow unhindered contact with the target of interest. Numerous chemical protocols to attach such ligands to liposomes and micelles have been developed over the past several years; it is utterly impossible even to briefly review those in a single book chapter. Here, we present a few representative protocols developed in our own laboratory that have been proven to be successful experimentally for generating a range of different types of lipid-based bioconjugates. The protocols described in the first sections of this chapter for the attachment of proteins/peptides/small molecules to liposomes and micelles involve modification of the phospholipid components with PEG (e.g., PEG-PE). Following this, the conjugation of the cationic

Design and Synthesis of Novel Functional Lipid-Based Bioconjugates

359

polymer polyethyleneimine (PEI) with lipids is described. Such types of lipid–polymer conjugates can produce stable complexes with negatively charged molecules, including DNA. In the final part of this chapter, a protocol for the modification of lipids with the chelating agent diethylenetriaminepentaacetic (DTPA) anhydride is described, which enables the conversion of nanocarriers (e.g., liposomes) into an efficient diagnostic tool for imaging applications.

2. Materials 2.1. Lipid Conjugation to Proteins/Peptides: Modification of Nanocarriers with mAb 2C5 or TATp via pNP-PEG3400-PE

1. pNP-PEG3400-pNP (bis(p-nitrophenyl) carbonate polyethylene glycol) (Laysan Bio, Inc., Arab, AL). Store at −20°C. 2. Lipids: 1,2-Dioleoyl-sn-glycero-3-phosphoethanolamine (DOPE); egg phosphatidylcholine (PC); cholesterol (Ch); and 1,2-dipalmitoyl-sn-glycero-3-phosphoethanolamine-N(lissamine rhodamine B sulfonyl) (Rh-PE) (Avanti Polar Lipids, Alabaster, AL). Store at −80°C. 3. Triethylamine (dry, in a septum-protected bottle) (Sigma, St. Louis, MO). 4. Sepharose® CL-4B medium (Sigma). 5. n-Octyl glucoside (OGP) (Sigma). 6. Fluorescein isothiocyanate (FITC)-dextran (4,400  Da) (Sigma). Store at 2–8°C. 7. Chloroform (dry, HPLC grade, no methanol). 8. Citrate-buffered saline (CBS): 5  mM Na-citrate, 141  mM NaCl, pH 5.0. Store at room temperature. 9. Phosphate buffer: 100  mM disodium phosphate heptahydrate, 0.72  mM monosodium phosphate monohydrate, pH 9.0. Store at room temperature. 10. Borate buffer: 0.1  M sodium tetraborate, 150  mM NaCl. Adjust the pH to 9.2 with 1N HCl or 1N NaOH. Store at room temperature. 11. 4-(2-Hydroxyethyl)-1-piperazine-ethanesulfonic acid (HEPES)-buffered saline (HBS): 5  mM HEPES, 141  mM NaCl, pH 7.4. Store at room temperature. 12. Tumor-specific monoclonal antinucleosome 2C5 antibody (mAb 2C5) (Harlan Laboratories, Indianapolis, IN). 13. Dialysis membrane tubing (2,000 Da and 250 kDa MWCO) (Spectrum Laboratories, Inc., Rancho Dominguez, CA). 14. TAT-peptide (TATp), 11-mer: Tyr-Gly-Arg-Lys-Lys-ArgArg-Gln-Arg-Arg-Arg (1,560 Da) (custom synthesized at the Tufts University Core Facility, Boston, MA).

360

Sawant and Torchilin

15. FreeZone freeze dry system (4.5  L) (Labconco, Kansas City, MO). 16. Mini-extruder (Avanti). 17. Bio-Gel A-1.5m gel filtration column (0.7 × 25 cm) (Bio-Rad). 18. Tris-buffered saline, (TBS): 50  mM Tris–HCl, 150  mM NaCl, pH 7.4. Store at room temperature. 19. TBS containing 0.05% (w/v) Tween 20 (TBST). Store at 2–8°C. 20. TBS containing 0.05% (w/v) Tween 20 and 2 mg/mL casein (TBST-Cas). Store at 2–8°C. 21. 40 mg/mL Poly-l-lysine (PLL, 30–70 kDa) solution in TBS. Store at 2–8°C. 22. 40  mg/mL Nucleosome solution in TBST-Cas. The nucleosomes are obtained from the water-soluble fraction of calf thymus nucleohistone (Worthington Biochemical Corp., Lakewood, NJ). Store at 2–8°C. 23. 10 mg/mL mAb 2C5 in TBST-Cas. Store at 2–8°C. 24. Goat anti-mouse IgG-horseradish peroxidase (HRP) conjugate (ICN Biomedicals, Aurora, OH) in TBST-Cas. The concentrated stock solution is stored at −20°C. Prepare a fresh working solution by diluting the stock solution 1:5,000 (v/v) in TBST. 25. Enhanced K-blue® TMB peroxidase substrate (Neogen, Lexington, KY). Store at 2–8°C. 26. Multiscan MCC/340 microplate reader equipped with GENESIS-LITE software (Thermo Labsystems, Franklin, MA). 2.2. Lipid Conjugation to Small Molecules

1. PEG-lipids: 1,2-Distearoyl-sn-glycero-3-phosphoethanolamineN-[methoxy (poly(ethylene glycol))-2000] (PEG2000-PE); 1,2-distearoyl-sn-glycero-3-phosphoethanolamine-N[amino(polyethylene glycol)2000] (PE-PEG2000-Amine) (Avanti Polar Lipids). Store at −80°C. 2. l-Ascorbic acid (Sigma). 3. Hydrogen bromide (HBr), 33% (w/w) solution in glacial acetic acid (Sigma). 4. Diethyl ether. 5. Dimethylformamide (DMF). 6. Sephadex® G25 superfine medium (Sigma). 7. Nuclear magnetic resonance (NMR) spectrometer (500 MHz) (Varian Inc., Palo Alto, CA). 8. Matrix-assisted laser desorption/ionization-time of flight (MALDI-TOF) mass spectrometer.

Design and Synthesis of Novel Functional Lipid-Based Bioconjugates

2.3. Lipid Conjugation to Polymers

1. 1-Palmitoyl-2-azelaoyl-sn-glycero-3-phosphocholie Ester) (Avanti Polar Lipids). Store at −80°C.

361

(azPC

2. Branched polyethyleneimine (bPEI) (1,800 Da) (Polysciences, Inc., Warrington, PA). 3. N,N ¢-Carbonyldiimidazole (CDI) (Fluka Chemie GmbH). 4. Plasmid DNA (pGFP) encoding green fluorescent protein (GFP) (1 mg/mL) (Elim Biopharmaceuticals, Hayward, CA). 5. HEPES-buffered glucose (HBG): 10 mM HEPES, 5% (w/v) d-Glucose, pH 7.4. 6. E-Gel electrophoresis system (Invitrogen/Life Technologies). 2.4. Liposome Conjugation to Metal-Loaded Diagnostic Reporters for Magnetic Resonance Imaging

1. N-Glutaryl phosphatidylethanolamine (NGPE) (Avanti Polar Lipids). Store at −80°C. 2. N,N ¢-Carbonyldiimidazole (Fluka Chemie GmbH). 3. N-Hydroxysuccinimide (Sigma). 4. Carbobenzoxy-poly-l-lysine (CBZ-PLL) (5,400 Da) (Sigma). 5. Triethylamine (dry, in a septum-protected bottle) (Sigma). 6. Diethylenetriaminepentaacetic (DTPA) anhydride (Sigma). 7. Succinic anhydride (Sigma). 8. Gadolinium(III) (Sigma).

chloride

hexahydrate

(GdCl 3 ⋅ 6H2O)

9. 5  MHz NMR proton spin analyzer (RADX Corporation, Houston, TX). 10. Sonic Dismembrator 60 probe tip sonicator (Fisher Scientific).

3. Methods 3.1. Lipid Conjugation to Proteins/Peptides: Modification of Nanocarriers with mAb 2C5 or TATp via pNP-PEG3400-PE

Although antibodies are the most commonly conjugated proteins, the same conjugation techniques can be applied to other types of proteins as well. Various coupling chemistries have been reported in the literature for the conjugation of proteins to lipids or to the PEG terminus of PEGylated liposomes or micelles. Indeed, several types of lipopolymers of the general formula X-PEG-PE, where “X  ” represents a reactive functional group-containing moiety and “PEG-PE” represents a conjugate of PEG and phosphatidylethanolamine (PE), are now commercially available and can be used for bioconjugation applications. Some examples of end-group functionalized PEG-lipids are given in Table 1. The most efficient and commonly used types of selective reactions for conjugating lipids to proteins include the following: In the first approach, activated carboxyl groups react with amino

362

Sawant and Torchilin

Table 1 Examples of end-group functionalized lipid–polymer conjugates of the general formula X-PEG-PE used in this work for the preparation of ligand-modified liposomes Functional group (X ) structure

Functional group (X ) name

Example applications of end-group functionalized lipid–polymer conjugates

H2N–

Primary amine

Amino-PEG-PE for preparation of long-circulating liposomes (e.g., end-group functionalization (11, 12) and synthesis of ligand-PEG-DSPE (13, 14) via the amino group modification)

HO2C–

Carboxylic acid

Modification and conjugation reactions via carbodiimidemediated coupling for the preparation of immunoliposomes (15, 16)

Hydrazide

Conjugation of antibodies containing oxidized carbohydrate residues (17, 18) to liposomes modified with PEG-hydrazide chains

Ortho-pyridyldisulfide PDP-PEG-DSPE conjugation to thiol-containing ligands (PDP) through disulfide bonds; also useful for preparation of the precursor to HS-PEG-DSPE (for attachment of maleimide (12) and bromoacetyl (19)-containing ligands) Maleimido

Maleimido-PEG-PE conjugation to Fab¢ fragments and other thiol-containing ligands (11, 20–23)

Carbonate

Conjugation of nitrophenyl carbonates (R = nitrophenyl) to amino-containing ligands (24–27)

groups to produce amide bonds. In the second type of reaction, pyridylthiols react with thiols to yield disulfide bonds. In the third method, maleimide derivatives react with thiols to generate thioether bonds. Finally, in the fourth case, a hydrazine group reacts with an oxidized carbohydrate group of the antibody (11, 15, 17, 28–32). For the preparation of ligand-bearing micelles or PEGylated liposomes, three types of approaches are commonly employed. In the first protocol, the end-group functionalized PEG-lipids (see Table 1 for examples) are first directly incorporated into the

363

Design and Synthesis of Novel Functional Lipid-Based Bioconjugates

O C

H

O

O CO

O

-N

O

-C

NH2-Ligand O

C

nd

ga

li H-

O

NO2

NO

2

nanocarriers and then conjugated to specific ligands (this approach is generally used for biomacromolecular ligands, such as immunoglobulins). However, in this approach it is possible that some of the reactive end-groups will remain on the surface of the nanocarrier, which may then crosslink through multiple attachments to a single protein molecule. It is also possible that some reactive groups on the inner surface of the liposomes will initially remain unreacted, and then subsequently undergo side reactions with drug molecules or other lipid components. Consequently, quenching of the unreacted end-groups is sometimes required (11). In the second protocol, the pure ligand–PEG-PE conjugate is synthesized first and then mixed with other liposome- or micelle-forming components. However, in the case of liposomes it is possible that slightly less than half of the synthesized conjugate molecules are oriented with the ligands facing the inner aqueous compartment,and thus are unavailable to undergo interactions with the target of interest (19). Finally, the third protocol, referred to as the “post-insertion technique” (33), involves the incubation of presynthesized ligand–PEG-PE conjugates with preformed plain or PEGylated liposomes (19). This process ensures that the ligand moieties are positioned on the outer surface of the liposomes. In this section, we describe a simple one-step procedure that can be used for binding a large variety of amino group-containing ligands (including antibodies and other proteins, peptides, and small molecules) to the distal end of a nanocarrier-attached (p-nitrophenyl) carbonyl-PEG-PE (pNP-PEG-PE) (24, 34). pNP-PEG-PE can be readily incorporated into liposomes and micelles via its phospholipid moiety, and it reacts easily with any amino group-containing substrate compound via its waterexposed pNP group to form a stable and nontoxic carbamate bond (Fig.  2). The reaction between the pNP group and the

pH 8-9.5

O NO

2

pNP-PEG-PE PEG-PE

Fig. 2. Schematic representation of ligand attachment to micelles via pNP-PEG-PE unimers.

OH

364

Sawant and Torchilin

amino group of the ligand proceeds easily and quantitatively at pH 8.0, while excessive free pNP groups are readily eliminated by spontaneous hydrolysis. In the example protocols presented below, the synthesis of pNP-PEG-PE is first described. This conjugate is then employed for the modification of liposomes with an antibody protein (viz., mAb 2C5) using the post-insertion technique. Following this, an alternative protocol for the modification of pNP-PEG-PE-bearing micelles with mAb 2C5 is also presented. And finally, in the last protocol of this section, we describe how peptides such as transactivating transcriptional activator peptide (TATp) can also be conjugated to liposomes containing PEG-PE using the same conjugation technique. 3.1.1. Synthesis of pNP-PEG3400-PE

1. Dissolve 800  mg of pNP-PEG3400-pNP in 10  mL of dry CHCl3 (80 mg/mL) in a 25 mL pear-shaped flask. 2. Add 33  mg of DOPE to the above solution, followed by 20 mL of triethylamine. Incubate the reaction mixture overnight at room temperature under argon (Ar) with stirring (see Note 1). 3. Remove the solvents on a rotary evaporator, followed by freeze-drying overnight to eliminate any residual solvents. 4. To purify the conjugates by size-exclusion (gel filtration) chromatography using gravity-feed (see step 6), prepare a column (e.g., 25 × 500 mm) that contains ~220 mL of clean Sepharose® CL-4B media equilibrated in 0.001 M HCl (pH 3). 5. Remove the dry reaction flask (from step 3) from the freezedryer. Add 4 mL of 0.001 M HCl (pH 3.0) to hydrate the PEG-lipid film (at a total PEG concentration of ~200 mg/mL) (see Note 2). 6. Apply the product solution (4  mL) obtained from step 5 above to the gel filtration column prepared in step 4, and elute with 0.001 M HCl (pH 3) (maximum elution volume =  300  mL). The product loading onto the column is thus approximately 1.8% (v/v) and 0.4% (w/v) (see Note 3). Collect the eluate using an automatic fraction collector containing appropriate-sized test tubes (e.g., 13 × 100 mm), and collect 75 drops (~4 mL) per fraction. 7. Perform a thin-layer chromatography (TLC) analysis to confirm the presence of the pNP-PEG3400-PE product in the collected fractions [use one-half of a 20 × 20 cm aluminum-backed sheet with a 10-min development time in chloroform/methanol/water, 80/20/2 (v/v/v)]. Once it is confirmed that all the product has eluted from the column, no more fractions need to be collected.

Design and Synthesis of Novel Functional Lipid-Based Bioconjugates

365

8. Freeze-dry all the fractions that contain the pNP-PEG3400-PE product. Store the product as a powder in amber vials, or dissolve it in CHCl3 at a concentration of 10 mg/mL (total PEG concentration) and store at −20 or at −80°C. 3.1.2. Modification of Liposomes with mAb 2C5 via the Post-Insertion Technique

1. Add a 40-fold molar excess of pNP-PEG3400-PE dispersed in a 10-mg/mL solution of octyl glucoside in 5 mM CBS (pH 5.0) to an equal volume of a 1 mg/mL solution of protein (e.g., mAb 2C5 or other IgG) in phosphate buffer, pH 9.0. Adjust the pH of the mixture to pH 8.5 with 1 M NaOH to facilitate the reaction between the (antibody) protein’s primary aminogroups and the pNP groups of pNP-PEG3400-PE. 2. Incubate the reaction mixture (pH 8.5) for 24  h at 4°C. Separate the micelles of lipid-conjugated mAb 25 from other reaction products by size-exclusion chromatography using a Sepharose® CL-4B column. 3. Mix the required amount of the purified mAb 2C5-PEG-PE conjugate (obtained from step 2 above) with a solution containing preformed liposomes (see Note 4) and incubate for 8 h at 4°C. 4. Separate the mAb 2C5-modified liposome product from the remaining octyl glucoside and free, nonincorporated antibodies by dialysis using cellulose ester membrane dialysis tubing (250 kDa MWCO).

3.1.3. Modification of Micelles with mAb 2C5 Using MicelleIncorporated pNP-PEG-PE

1. To attach the mAb 2C5 protein to micelles to obtain immunomicelles, supplement a chloroform solution of PEG-PE with 5 mol% of the reactive component, pNP-PEG-PE, in a round-bottom flask (see Note 5). 2. Remove the organic solvents by rotary evaporation, followed by freeze-drying. 3. Hydrate the dried film with 5 mM CBS (pH 5.0) to obtain a final concentration of 20 mg/mL of total lipid. 4. To 0.5 mL of the resultant mixture, add 0.3 mL of a 2.94mg/mL mAb 2C5 solution in 100  mM phosphate buffer (pH 9.0) with vortexing. Adjust the pH of the mixture to 8.5 with 1 M NaOH to facilitate the reaction between the (antibody) protein’s primary amino-groups and the pNP groups of pNP-PEG-PE to yield immunomicelles. 5. Incubate the reaction mixture for 3 h at room temperature. 6. Dialyze the immunomicelles against 1 L of 5 mM HBS, pH 7.4 using cellulose ester membrane dialysis tubing (250 kDa MWCO).

3.1.4. Modification of Liposomes with TATp Using pNP-PEG-PE

1. Prepare a lipid mixture containing PC, Ch, and pNP-PEG-PE (7/3/0.05, molar ratio) in chloroform in a round-bottom flask (see Note 6). (Optional: For visualization, label the

366

Sawant and Torchilin

­liposomes by adding 0.5 mol% of Rh-PE here in step 1, or add FITC-dextran (45 mg/mL) to the CBS solution later in step 3.) 2. Remove the solvents by rotary evaporation, followed by freeze-drying. 3. Rehydrate the film in 5 mM CBS, pH 5.4 (see Note 7). 4. To form the liposomes, extrude the lipid dispersion 21 times through polycarbonate filters (pore size 200 nm) with a Miniextruder apparatus. 5. If FITC-dextran (45 mg/mL) was added to the CBS solution during hydration (step 3), remove nonincorporated FITCdextran on a Bio-Gel A-1.5m gel filtration column (0.7 × 25 cm). 6. To attach TATp to the liposomes, add 1 mg of the peptide in borate buffer to 10  mg of pNP-PEG-PE-containing liposomes in CBS (see Note 8) and incubate overnight at room temperature. Purify the TATp-liposome product from unbound TATp and released pNP by size-exclusion chromatography on a Bio-Gel A-1.5m column (0.7 × 25 cm). 7. Alternatively, prepare TATp-PEG3400-PE first by reacting TATp with pNP-PEG3400-PE as follows, and then add it to the lipid mixture in step 1 instead of pNP-PEG-PE. In a typical case, incubate a solution of 5 mg of TATp in 1 mL of chloroform supplemented with 10 mL of triethylamine with 12 mg of pNP-PEG3400-PE in 0.6 mL of chloroform overnight with stirring at room temperature. Remove the organic solvents by rotary evaporation and freeze-drying. To the resultant film, add 1 mL of deionized water and vortex to remove the film from the flask. Purify the TATp-PEG3400-PE product away from unconjugated TATp and released pNP by dialyzing the solution against water at room temperature using a cellulose ester membrane dialysis tubing (2,000 Da MWCO). 3.1.5. Analysis of the Activity of mAb 2C5-Modified Micelles by ELISA

1. To a microplate, add 50 mL/well of a 40-mg/mL PLL solution in TBS and incubate overnight at 4°C. 2. Discard the PLL solution and add 200 mL of TBST-Cas block solution to each well. Incubate the block solution in the microplate for 1 h at room temperature (to prevent nonspecific binding during assays). 3. Discard the TBST-Cas solution and wash the wells three times with 200 mL of TBST. 4. To each well, add 50 mL of a 40-mg/mL nucleosome solution in TBST-Cas and incubate the microplate for 1  h at room temperature.

Design and Synthesis of Novel Functional Lipid-Based Bioconjugates

367

0.60

Absorbance 620/492 nm

0.50 0.40 0.30

Native mAb 2C5 2C5-Paclitaxel-micelles

0.20

IgG-Paclitaxel micelles Plain-Micelles

0.10 0.00 0

2

4

6

8

10

12

−0.10

Concentration of protein (µg/mL) Fig. 3. Immunoreactivity of 2C5-PEG-PE micelles as determined by ELISA.

5. Discard the nucleosome solution and wash the wells three times with 200 mL of TBST. 6. Incubate the wells with varying concentrations of native mAb 2C5 protein or mAb 2C5-modified micelles for 1 h at room temperature. 7. Discard the samples and wash the wells three times with 200 mL of TBST. 8. To each well, add 50 mL of a 1:5,000 (v/v) dilution of goat anti-mouse IgG peroxidase conjugate in TBST-Cas. Incubate the microplate for 1 h at room temperature. 9. Discard the goat anti-mouse IgG peroxidase conjugate solution and wash the wells three times with 200 mL of TBST. 10. To each well, add 100 mL of enhanced K-blue® TMB peroxidase substrate and incubate for 15 min. 11. Read the absorbance using a Labsystems Multiscan MCC/340 microplate reader equipped with a dual-wavelength detector at l = 620 nm (reference filter l = 492 nm). 12. Representative results obtained for an ELISA using various synthesized immunomicelles are shown in Fig. 3. 3.2. Lipid Conjugation to Small Molecules

The attachment of small molecules such as folate to the surface of nanocarriers is extremely important for targeting therapeutic drugs to cells expressing appropriate receptors. Various chemistries have been reported in the literature for the preparation of folate-modified liposomes (35–37). In addition, the coupling of p-aminophenyl-sugar derivatives to NGPE-containing liposomes

368

Sawant and Torchilin

has also been described for the modification of lipid-based nanocarriers with sugar moieties. Any amino group-containing small molecule can be easily conjugated using the pNP-PEG-PE polymer as described earlier in Subheading 3.1. In this section, we describe a method developed in our laboratory for the synthesis of 6-ascorbate-PEG-PE conjugates that can be incorporated into nanocarriers. These ascorbate-modified nanocarriers have been shown to be useful for brain cell targeting in vitro (38). As outlined in detail below, the 6-ascorbate-PEG-PE is synthesized using a two-step procedure: (1) activation of ascorbic acid with bromine to produce 6-Br-ascorbic acid and (2) synthesis of 6-ascorbate-PEG-PE by reacting PE-PEG2000-Amine with an excess of 6-Br-ascorbic acid. 3.2.1. Synthesis of 6-Bromo-6-DeoxyAscorbic Acid

1. Add 2  g of ascorbic acid (11.35  mmol) to 2.7  mL of 33% (w/v) HBr in glacial acetic acid (17.18  mmol) and stir the suspension overnight at room temperature. 2. Remove the HBr and acetic acid by nitrogen (N2) flux and rotary evaporation. 3. Add 5 mL of deionized water to the syrup, and stir this mixture for 30 min at 60°C until the volume is reduced by one-half. 4. Extract the resulting residue six times with 5  mL of ethyl acetate, and remove traces of water by sodium sulfate treatment. 5. Remove the organic solvent under vacuum, and dissolve the product in 3 mL of acetonitrile prewarmed to 70°C. 6. Cool the solution to 4°C to allow crystallization of the product. Redissolve the white crystals and then recrystallize the product according to the same procedure. 7. Recover the final product by filtration and dry under vacuum overnight. 8. Dissolve the recovered 6-bromodeoxy-ascorbic acid product in DMSO-d6 and analyze by 1H NMR and 13C NMR using a Varian 500  MHz spectrometer. 1H NMR: d 3.36 (dd, 1H, J = 10.06, d 7.14 Hz, BrCCHH), d 3.60 (dd, 1H, J = 10.06, 6.56  Hz, BrCCHH ), d 3.96 (t, 1H, J = 6.67  Hz, BrCH2CHOH), d 4.83 (broad s, 1H, OCH), d 5.63 (broad s, 5.63, 1H, BrCH2CHOH ), d 8.42 (s, 1H, OCOCOHCOH ), d 11.66 (s, 1H, OCOCOHCOH). 13C NMR: d 34.6 (BrC), 68.2 (BrCC), 75.2 (OCH), 152.3 (OCCOH), 118.1 (OCCOH=COH), 170.3 (OC=O). 9. Analyze the 6-bromodeoxy-ascorbic acid product by MALDI-TOF mass spectrometry using 2,5-dihydroxybenzoic acid as the sample matrix. The negative ion is found at m/z 237.05.

Design and Synthesis of Novel Functional Lipid-Based Bioconjugates 3.2.2. Synthesis of 6-Ascorbate-PEG2000-PE

369

1. Add 2.8 mL of a 25 mg/mL solution of PE-PEG2000-Amine in chloroform (25 mmol, 70 mg) into a 25-mL round-bottom flask. 2. Remove the solvent by rotary evaporation, and dissolve the resulting solid residue in 2 mL of DMF. 3. Add 60 mg of 6-bromodeoxy-ascorbic acid (250 mmol) and 35 mL of triethylamine (250 mmol) and stir the solution overnight under argon (Ar) in the dark. 4. Precipitate the mixture by dropwise addition of cold diethyl ether while stirring. Wash the crude precipitate several times with diethyl ether, and dry the solid under vacuum. 5. Dissolve the dry powder in 2 mL of deionized water. Purify the product by size-exclusion chromatography using a column prepacked with Sephadex® G25 superfine medium, and elute with deionized water. Fractions that are positive by both UV analysis and an iodine test are pooled together and lyophilized. The product yield is 40 mg (57%). 6. In order to assess the PEG/ascorbate molar ratio in the 6-ascorbate-PEG-PE conjugate, an iodine test for PEG (see Note 9), a UV spectroscopic analysis for ascorbate (see Note 10), and a Snyder colorimetric test for the residual unconjugated PEG amino-groups (see Note 11) are carried out on a weighed amount of the purified derivative. 7. Dissolve the recovered 6-ascorbate-PEG-PE product in CDCl3 and analyze by 1H NMR spectroscopy. 1H NMR: d 3.73 (d, 1H, OCHCHOHCH2 of ascorbate), d 3.642 (s, ~180 H, –[OCH2CH2]n– of PEG), d 2.308 (m, 4H, OCOCH2 of phospholipid stearate), 1.249 (m, 56 H, –[CH2]n– of phospholipid stearate).

3.2.3. Modification of Nanocarriers with 6-Ascorbate-PEG-PE

The modification of nanocarriers with 6-ascorbate-PEG-PE can be achieved by incorporating the 6-ascorbate-PEG-PE conjugate component at the desired mol% in a lipid mixture during the preparation of micelles and liposomes, as described previously in Subheading 3.1.

3.3. Lipid Conjugation to Polymers

Among nonviral vectors, the cationic polymer PEI and its derivatives have been widely explored in gene delivery research (39–42). PEI has the distinct advantage of possessing the highest positive charge density among synthetic polycations, which enables effective condensation of DNA by electrostatic interactions. PEI is also capable of mediating endosomal escape by the “proton sponge” mechanism (40, 41) and nuclear localization (43), which allows for high transfection efficiencies. PEI is available in a wide range of molecular weights from approximately 1–800 kDa, and may be obtained in either linear or branched forms. Low-molecular weight PEI

370

Sawant and Torchilin

has been shown to be well-tolerated by eukaryotic cells with low toxicity (44). However, the in  vivo application of PEI/DNA complexes is limited because of rapid clearance from circulation and accumulation in reticuloendothelial system (RES) organs such as the liver and spleen (45). When injected systemically, PEI/DNA complexes are also subject to DNA dissociation and aggregation effects (45). Several approaches have been proposed to provide PEI/DNA complexes with a higher level of in  vivo stability (39, 42, 46). Lipid-grafted PEI conjugates such as cetylated PEI (47) and cholesteryl PEI (48) have been used to prepare polycationic liposomes (PCL) loaded with DNA. For example, a water-soluble lipopolymer (WSLP) consisting of a low-molecular weight PEI and cholesterol was employed for in vivo gene therapy for cancer treatment and for gene delivery to ischemic myocardium (49–51). In other work, preformed PEI/DNA complexes were encapsulated in PEG-stabilized liposomes, resulting in so-called “precondensed stable plasmid lipid particles” (pSPLPs) (52). In our own laboratory, we have synthesized a novel phospholipid-PEI conjugate (PLPEI) (described below) that in combination with other unmodified lipids and PEG-PE can self-assemble into monolayerenveloped hard-core micelle-like nanoparticles (MNPs) in the presence of plasmid DNA. In a recent report, we demonstrated how the resulting MNPs have architectures and properties suitable for in vivo use (53). 3.3.1. Synthesis of Phospholipid– Polyethyleneimine Conjugate

1. Add 12 mg of bPEI (7 mmol) to 0.5 mL of chloroform and mix with 5 mg of oxidized PC (azPC Ester, 7 mmol) dissolved in 1 mL of chloroform. Assuming that bPEI has a 1:2:1 molar ratio of primary:secondary:tertiary amine groups, the resulting reaction mixture will correspond to an acid group-to-primary amine group molar ratio of about 1:10 (i.e., the polymer contains an excess of reactive amines). 2. Add 0.5 mg of CDI (3 mmol) to the above solution to activate the acid groups on the azPC Ester by forming an imidazolide derivative. Incubate the reaction mixture with 10 mL of triethylamine at room temperature for 24 h with stirring. 3. Remove the chloroform under a stream of N2 gas and resuspend the residue with 2 mL of deionized water. 4. Purify the product by dialysis against deionized water using membrane dialysis tubing (2,000  Da MWCO), and then freeze-dry. 5. Dissolve the purified PLPEI product in CDCl3 and analyze by 1H NMR spectroscopy. The extent of conjugation is determined to be a 1:1 molar ratio of PEI to lipid from the ratio of the ethylene (–CH2CH2–) signal (2.4–2.8 ppm) of the PEI main chain to the methyl (–CH3) signal of the phospholipid

Design and Synthesis of Novel Functional Lipid-Based Bioconjugates

371

Fig. 4. Agarose gel electrophoresis of PLPEI/DNA complexes (top panel ) and PEI/DNA complexes (bottom panel ) at varying N/P ratios.

head (3.4 ppm) in the resulting NMR spectrum (d 0.9:2.7H, d 1.3:17.6H, d 1.6:5.4H, d 2.4–2.8:96.0H, d 3.3:12.8H, d 3.6:1.58H, d 4.0–4.6:5.43H). 3.3.2. Complexation of Plasmid DNA with PLPEI and Analysis by Electrophoresis

A fixed amount of plasmid DNA (pGFP; 100 mg) encoding green fluorescent protein and varying amounts of PLPEI are separately diluted in HBG solution to a final volume of 250 mL. The PLPEI solution is then transferred to the DNA solution by rapid addition and vortexing. The resulting PLPEI/DNA polyplexes are analyzed by agarose gel electrophoresis using an E-Gel electrophoresis system. Representative results from an agarose gel electrophoresis analysis of PLPEI/DNA complexes at varying amineto-phosphate (N/P) ratios are shown in Fig. 4 (see Note 12).

3.4. Liposome Conjugation to Metal-Loaded Diagnostic Reporters for Magnetic Resonance Imaging

Medical diagnostic imaging techniques such as gamma scintigraphy, computed tomography (CT), ultrasonography, and magnetic resonance imaging (MRI) require a sufficient level of signal intensity from an area of interest in order to differentiate it from surrounding tissues. To increase the local spatial concentration of an imaging contrast agent, various types of particulate carriers that are able to carry multiple contrast moieties to a given target area of interest have been exploited. Liposome-based contrast agents for MRI act by shortening the relaxation times (T1 and T2) of the surrounding water protons, resulting in an increase (T1 agents) or a decrease (T2 agents) in the intensity of tissue signals (54–56). Because of the toxicity and poor solubility of free paramagnetic heavy metal cations under physiological pH conditions, chelated metal complexes are used in most MRI-T1 contrast agent designs (57). Membranotropic chelating agents typically consist

372

Sawant and Torchilin

of a polar head group containing the chelated paramagnetic atom and a lipid moiety that anchors the metal-chelate complex in the liposomal membrane, and thereby provide better relaxivity values than those obtained for liposome-encapsulated paramagnetic ions. Some examples of such types of chelating agents include DTPA-stearylamine (DTPA-SA) (58), DTPA-phosp­hatidylamine (DTPA-PE) (59–61), and DTPA-bis(methyl­amide) (DTPA-BMA) (62, 63). To further increase the signal intensity per liposome, we have developed in our laboratory poly-l-lysine (PLL)-based polychelating amphiphilic polymers (PAP) (64). Such polymers can be loaded with multiple metal atoms via main chain-attached multiple chelating groups, and firmly incorporated into liposomal membranes via a lipid residue at one terminus of the molecule. The advantage of this approach is that a single lipid anchor can carry a polymer molecule with multiple metal-loaded chelating side-groups on the liposome surface (65, 66). 3.4.1. Synthesis of Polychelating Amphiphilic Polymer DTPA-PLL-NGPE

1. Activate 25 mg of NGPE with 20 mg of N,N¢-carbonyldii­ midazole in the presence of 13 mg of N-hydroxysuccinimide (NHS) for 16 h at room temperature. 2. Add 186  mg of CBZ-PLL and 5  mL of triethylamine, and then continue to incubate the reaction for another 5  h at room temperature with stirring. After approximately 1 h, add an additional 200 mL of chloroform (see Note 13). 3. Dry the reaction mixture on a rotary evaporator. Suspend the recovered CBZ-PLL-NGPE product in ~25 mL of deionized water, filter, and then freeze-dry. 4. Dissolve 174 mg of CBZ-PLL-NGPE in 8 mL of HBr solution (33% w/w in glacial acetic acid) and allow the reaction to proceed for 2 h at room temperature. 5. Precipitate the deprotected PLL-NGPE product with ~20 mL of dry ether. Wash the precipitated product further with dry ether and then freeze-dry. 6. Suspend 150  mg of PLL-NGPE in ~2  mL of chloroform/ methanol (1:1, v/v) and react with 559  mg of DTPA anhydride in 2 mL of DMSO in the presence of 200 mL of triethylamine for 16 h at room temperature with stirring. 7. Add 485 mg of succinic anhydride in 1 mL of DMSO to block any remaining polymer amino-groups and continue to incubate the reaction for an additional 1 h at room temperature. 8. Purify the reaction mixture from water-soluble compounds by dialysis against deionized water (3,500 Da MWCO) (see Note 14). 9. Freeze-dry the purified solution of the final DTPA-PLLNGPE product.

1/ T1 (1/msec)

Design and Synthesis of Novel Functional Lipid-Based Bioconjugates 0.08 0.07 0.06 0.05 0.04 0.03 0.02 0.01 0.00

Gd-DTPA-PLL-NGPE

0

1

373

Gd-DTPA-PE

2 Lipid Concentration (mM)

3

4

Fig. 5. Molecular relaxivity (1/T1) of liposomes containing the same molar fractions of Gd-DTPA-PE and Gd-DTPA-PLLNGPE (n = 6).

3.4.2. Loading of DTPA-PLL-NGPE with Gd Ions

1. Add 150  mg of GdCl3⋅6H2O in 0.25  mL of 0.1  M citrate buffer (pH 5.3) to 25 mg of DTPA-PLL-NGPE suspended in ~2 mL of dry pyridine. Incubate the reaction mixture for 2 h at room temperature with stirring. 2. Dialyze the mixture against deionized water (3,500  Da MWCO) and freeze-dry (see Notes 14 and 15). 3. Use the freeze-dried Gd-DTPA-PLL-NGPE conjugate product without further purification to synthesize liposomes containing the desired lipid formulation (for example, see Subheading 3.4.3). 4. Measure the in vitro relaxation parameters of the synthesized liposomes using a RADX 5 MHz NMR proton spin analyzer at room temperature in HBS (pH 7.4) at different liposome concentrations. Representative results for a relaxivity analysis of liposomes containing Gd-DTPA-PLL-NGPE or a monomeric chelate (Gd-DTPA-PE) (66) are shown in Fig. 5.

3.4.3. Preparation of Liposomes Containing Gd-DTPA-PLL-NGPE

1. Prepare a lipid mixture containing PC, Ch (2/1, molar ratio), 5 mol% PEG2000-PE and 1.75 mol% of Gd-DTPA-PLL-NGPE or Gd-DTPA-PE in chloroform in a round-bottom flask. 2. Remove the solvents by rotary evaporation, followed by freeze-drying. 3. Rehydrate the film in a 50-mg/mL solution of octyl glucoside in 5 mM HBS, pH 7.4. 4. Homogenize the lipid mixture by sonication at ~7  W for 30 min using a probe tip sonicator (Sonic Dismembrator 60, Fisher Scientific). 5. To remove metal traces derived from the tip of the probe sonicator and to separate larger lipid aggregates from the liposomes, centrifuge the liposome sample for 10 min at ~3,000 × g. 6. To form the liposomes, remove the octyl glucoside by dialyzing the lipid mixture against HBS.

374

Sawant and Torchilin

4. Notes 1. Monitor the reaction by thin-layer chromatography (TLC; stationary phase: Silica Gel 60 F254 TLC plate; mobile phase: chloroform/methanol/water, 80/20/2 v/v/v). Use UV light for general visualization. For specific visualization, use Dragendorff’s reagent for identifying PEG spots and use molybdenum reagent for identifying PE spots (phosphorus detection). Compare the reaction spots with the appropriate standards; the total elimination of the PE spot from a reaction sample and the appearance of a combined PEG/PE spot will indicate a complete reaction. For TLC spotting, diluting the reaction mixture aliquot to a final PEG concentration of less than 10 mg/mL will help to distinguish the TLC spots. 2. Shake and swirl the flask by hand, and then vortex until dissolution of the pNP-PEG3400-PE film occurs. The PEG-PE derivative and unreacted PEG must be fully dispersed within the loading solution for separation to occur during the gel filtration chromatography step. If the product is found to be difficult to dissolve, the solution can be carefully sonicated (e.g., in a water bath for not more than 5 min at 15 W). 3. Do not activate the fraction collector until about 25% of the column volume has passed (~55 mL). The first ~55 mL (holding volume) is collected separately in a beaker. 4. As an example, we have used commercially available Doxil® liposomes (Centocor Ortho Biotech, Horsham, PA) for conjugation reactions with the 2C5 antibody. 5. Typically, we observe that about 30% of the added mAb 2C5 protein becomes attached to micelles containing 2 mol% of pNP-PEG-PE. From this yield value, it can be calculated that up to ten antibody molecules bind to a single micelle. Protein binding to micelles without pNP-PEG-PE is negligible. The antibody attachment yield may be increased by increasing the molar fraction of pNP-PEG-PE in the micelles. The yield may reach as high as 50% when micelles contain 8 mol% of pNPPEG-PE. Excessive amounts of pNP-PEG-PE, however, may cause over-modification of a protein molecule and thus lead to its inactivation. 6. pNP-PEG-PE is capable of incorporation into both liposomes and micelles via its hydrophobic PE group and easily attaches to any amino group-containing ligand via the pNP group with the formation of a stable carbamate bond. As little as 0.5 mol% of pNP-PEG-PE incorporated into a lipid mixture provides a sufficient number of reactive groups on the liposomal surface to bind approximately 500 TATp molecules

Design and Synthesis of Novel Functional Lipid-Based Bioconjugates

375

per single 200-nm liposome (as estimated by following a TATp-associated radiolabel). 7. It is important to hydrate the film in acidic buffer conditions in order to prevent hydrolysis of the pNP groups. 8. After addition of borate buffer, the final pH of the mixture should be between 8.0 and 8.5 for efficient conjugation of the amino groups of the peptide with pNP-PEG-PE. 9. The iodine test for PEG is performed according to ref. 67: Prepare an iodine solution by dissolving 1.27  g of I2 in 100 mL of 2% (w/v) KI in water. Prepare a 5% (w/v) barium chloride (BaCl2) solution in 1N HCl. Prepare the PEG standard solutions and unknown sample solutions as follows: Add 500 mL of deionized water, 250 mL of BaCl2 solution, 250 mL of iodine solution, and 25  mL of sample (for the reagent blank, use 25 mL of deionized water). Incubate the solutions for 15  min at room temperature, and then read the absorbance at l = 535 nm against the reagent blank. 10. The UV spectroscopic analysis for ascorbate is performed according to ref. 68: Measure the ascorbate samples and unknown samples using a UV spectrophotometer at 264 nm. 11. The Snyder colorimetric test for residual unconjugated PEG amino-groups is performed according to ref. 69: Add 30 mL of sample solution (e.g., a solution of 2 mM mPEG-NH2; or the reaction mixture containing PEG-NH2 and the aminocontaining molecule “X,” diluted in water to give 2  mM PEG-NH-X) to 940 mL of 0.1 M borate buffer (pH 9.3) and 30 mL of aqueous TNBS (prepared by diluting the commercial solution of 5% TNBS fivefold with water). Incubate the reaction mixture for 15 min at room temperature. Read the absorbance at l = 420 nm against the reagent blank. 12. The desired N/P ratio can be calculated by assuming that each repeating unit of PEI (containing one amine group) has a molecular weight value of 43.1 g/mol, and that each repeating unit of DNA (containing one phosphate group) has a molecular weight value of 330 g/mol. 13. Conversion of the initial NGPE reagent is checked by TLC (Stationary phase: Silica Gel 60 F254 TLC plate; mobile phase: chloroform/methanol/water, 65/25/4 v/v/v). UV light is used for general visualization. The Rf value of the initial NGPE reagent = 0.37. The Rf value of the product N,a(e-CBZ-PLL) NGPE = 0.59. 14. Perform dialysis overnight with extensive changes of buffer. 15. The suspension may be viscous; in this case, it should be warmed.

376

Sawant and Torchilin

References 1. Lasic, D. D. (1993) Liposomes: From Physics to Applications, Elsevier, Amsterdam. 2. Torchilin, V. P. (2005) Recent advances with liposomes as pharmaceutical carriers. Nat Rev Drug Discov 4, 145–160. 3. Lasic, D. D., and Martin, F. J. (1995) Stealth Liposomes, CRC Press, Boca Raton. 4. Lukyanov, A. N., and Torchilin, V. P. (2004) Micelles from lipid derivatives of water-soluble polymers as delivery systems for poorly soluble drugs. Adv Drug Deliv Rev 56, 1273–1289. 5. Gao, Z., Lukyanov, A. N., Chakilam, A. R., and Torchilin, V. P. (2003) PEG-PE/phosphatidylcholine mixed immunomicelles specifically deliver encapsulated taxol to tumor cells of different origin and promote their efficient killing. J Drug Target 11, 87–92. 6. Mu, L., Elbayoumi, T. A., and Torchilin, V. P. (2005) Mixed micelles made of poly(ethylene glycol)-phosphatidylethanolamine conjugate and d-alpha-tocopheryl polyethylene glycol 1000 succinate as pharmaceutical nanocarriers for camptothecin. Int J Pharm 306, 142–149. 7. Roby, A., Erdogan, S., and Torchilin, V. P. (2006) Solubilization of poorly soluble PDT agent, meso-tetraphenylporphin, in plain or immunotargeted PEG-PE micelles results in dramatically improved cancer cell killing in vitro. Eur J Pharm Biopharm 62, 235–240. 8. Wang, J., Mongayt, D. A., Lukyanov, A. N., Levchenko, T. S., and Torchilin, V. P. (2004) Preparation and in vitro synergistic anticancer effect of vitamin K3 and 1,8-diazabicyclo[5,4,0]undec-7-ene in poly(ethylene glycol)-diacyllipid micelles. Int J Pharm 272, 129–135. 9. Lukyanov, A. N., Hartner, W. C., and Torchilin, V. P. (2004) Increased accumulation of PEG-PE micelles in the area of experimental myocardial infarction in rabbits. J Control Release 94, 187–193. 10. Torchilin, V. P. (2005) Lipid-core micelles for targeted drug delivery. Curr Drug Deliv 2, 319–327. 11. Kirpotin, D., Park, J. W., Hong, K., Zalipsky, S., Li, W. L., Carter, P., Benz, C. C., and Papahadjopoulos, D. (1997) Sterically stabilized anti-HER2 immunoliposomes: design and targeting to human breast cancer cells in vitro. Biochemistry 36, 66–75. 12. Allen, T. M., Brandeis, E., Hansen, C. B., Kao, G. Y., and Zalipsky, S. (1995) A new strategy for attachment of antibodies to sterically stabilized liposomes resulting in efficient targeting to cancer cells. Biochim Biophys Acta 1237, 99–108.

13. Wong, J. Y., Kuhl, T. L., Israelachvili, J. N., Mullah, N., and Zalipsky, S. (1997) Direct measurement of a tethered ligand-receptor interaction potential. Science 275, 820–822. 14. Gabizon, A., Horowitz, A. T., Goren, D., Tzemach, D., Mandelbaum-Shavit, F., Qazen, M. M., and Zalipsky, S. (1999) Targeting folate receptor with folate linked to extremities of poly(ethylene glycol)-grafted liposomes: in vitro studies. Bioconjug Chem 10, 289–298. 15. Maruyama, K., Takizawa, T., Yuda, T., Kennel, S. J., Huang, L., and Iwatsuru, M. (1995) Targetability of novel immunoliposomes modified with amphipathic poly(ethylene glycol)s conjugated at their distal terminals to monoclonal antibodies. Biochim Biophys Acta 1234, 74–80. 16. Blume, G., Cevc, G., Crommelin, M. D., Bakker-Woudenberg, I. A., Kluft, C., and Storm, G. (1993) Specific targeting with poly(ethylene glycol)-modified liposomes: coupling of homing devices to the ends of the polymeric chains combines effective target binding with long circulation times. Biochim Biophys Acta 1149, 180–184. 17. Hansen, C. B., Kao, G. Y., Moase, E. H., Zalipsky, S., and Allen, T. M. (1995) Attachment of antibodies to sterically stabilized liposomes: evaluation, comparison and optimization of coupling procedures. Biochim Biophys Acta 1239, 133–144. 18. Harding, J. A., Engbers, C. M., Newman, M. S., Goldstein, N. I., and Zalipsky, S. (1997) Immunogenicity and pharmacokinetic attributes of poly(ethylene glycol)-grafted immunoliposomes. Biochim Biophys Acta 1327, 181–192. 19. Zalipsky, S., Mullah, N., Harding, J. A., Gittelman, J., Guo, L., and DeFrees, S. A. (1997) Poly(ethylene glycol)-grafted liposomes with oligopeptide or oligosaccharide ligands appended to the termini of the polymer chains. Bioconjug Chem 8, 111–118. 20. Shahinian, S., and Silvius, J. R. (1995) A novel strategy affords high-yield coupling of antibody Fab¢ fragments to liposomes. Biochim Biophys Acta 1239, 157–167. 21. Park, K., Lee, M. Y., Kim, K. S., and Hahn, S. K. Target specific tumor treatment by VEGF siRNA complexed with reducible polyethyleneimine-hyaluronic acid conjugate. Biomaterials 31, 5258–5265. 22. Cheng, W. W., and Allen, T. M. (2008) Targeted delivery of anti-CD19 liposomal doxorubicin in B-cell lymphoma: a comparison of whole monoclonal antibody, Fab¢ fragments and single chain Fv. J Control Release 126, 50–58.

Design and Synthesis of Novel Functional Lipid-Based Bioconjugates 23. Mamot, C., Drummond, D. C., Greiser, U., Hong, K., Kirpotin, D. B., Marks, J. D., and Park, J. W. (2003) Epidermal growth factor receptor (EGFR)-targeted immunoliposomes mediate specific and efficient drug delivery to EGFR- and EGFRvIII-overexpressing tumor cells. Cancer Res 63, 3154–3161. 24. Torchilin, V. P., Levchenko, T. S., Lukyanov, A. N., Khaw, B. A., Klibanov, A. L., Rammohan, R., Samokhin, G. P., and Whiteman, K. R. (2001) p-Nitrophenylcarbonyl-PEG-PE-liposomes: fast and simple attachment of specific ligands, including monoclonal antibodies, to distal ends of PEG chains via p-nitrophenylcarbonyl groups. Biochim Biophys Acta 1511, 397–411. 25. Torchilin, V. P., Rammohan, R., Weissig, V., and Levchenko, T. S. (2001) TAT peptide on the surface of liposomes affords their efficient intracellular delivery even at low temperature and in the presence of metabolic inhibitors. Proc Natl Acad Sci U S A 98, 8786–8791. 26. Lukyanov, A. N., Elbayoumi, T. A., Chakilam, A. R., and Torchilin, V. P. (2004) Tumortargeted liposomes: doxorubicin-loaded longcirculating liposomes modified with anti-cancer antibody. J Control Release 100, 135–144. 27. Sawant, R. M., Cohen, M. B., Torchilin, V. P., and Rokhlin, O. W. (2008) Prostate cancerspecific monoclonal antibody 5D4 significantly enhances the cytotoxicity of doxorubicinloaded liposomes against target cells in vitro. J Drug Target 16, 601–604. 28. Sofou, S., and Sgouros, G. (2008) Antibodytargeted liposomes in cancer therapy and imaging. Expert Opin Drug Deliv 5, 189–204. 29. Torchilin, V. (2008) Antibody-modified liposomes for cancer chemotherapy. Expert Opin Drug Deliv 5, 1003–1025. 30. Sapra, P., Tyagi, P., and Allen, T. M. (2005) Ligand-targeted liposomes for cancer treatment. Curr Drug Deliv 2, 369–381. 31. Zalipsky, S. (1993) Synthesis of an end-group functionalized polyethylene glycol-lipid conjugate for preparation of polymer-grafted liposomes. Bioconjug Chem 4, 296–299. 32. Torchilin, V. P., Weissig, V., Martin, F. J., and Heath, T. D. (2003) Surface modifications of liposomes, in Liposomes: A practical approach (Torchilin, V. P., and Weissig, V., Eds.) pp 193–299, Oxford University Press, Oxford, New York. 33. Ishida, T., Iden, D. L., and Allen, T. M. (1999) A combinatorial approach to producing sterically stabilized (Stealth) immunoliposomal drugs. FEBS Lett 460, 129–133.

377

34. Torchilin, V. P., Lukyanov, A. N., Gao, Z., and Papahadjopoulos-Sternberg, B. (2003) Immunomicelles: targeted pharmaceutical carriers for poorly soluble drugs. Proc Natl Acad Sci U S A 100, 6039–6044. 35. Hilgenbrink, A. R., and Low, P. S. (2005) Folate receptor-mediated drug targeting: from therapeutics to diagnostics. J Pharm Sci 94, 2135–2146. 36. Pan, X., and Lee, R. J. (2007) Construction of anti-EGFR immunoliposomes via folate-folate binding protein affinity. Int J Pharm 336, 276–283. 37. Wu, J., Liu, Q., and Lee, R. J. (2006) A folate receptor-targeted liposomal formulation for paclitaxel. Int J Pharm 316, 148–153. 38. Salmaso, S., Pappalardo, J. S., Sawant, R. R., Musacchio, T., Rockwell, K., Caliceti, P., and Torchilin, V. P. (2009) Targeting glioma cells in  vitro with ascorbate-conjugated pharmaceutical nanocarriers. Bioconjug Chem 20, 2348–2355. 39. Kichler, A. (2004) Gene transfer with modified polyethylenimines. J Gene Med 6 Suppl 1, S3–10. 40. Boussif, O., Lezoualc’h, F., Zanta, M. A., Mergny, M. D., Scherman, D., Demeneix, B., and Behr, J. P. (1995) A versatile vector for gene and oligonucleotide transfer into cells in culture and in  vivo: polyethylenimine. Proc Natl Acad Sci U S A 92, 7297–7301. 41. Akinc, A., Thomas, M., Klibanov, A. M., and Langer, R. (2005) Exploring polyethylenimine-mediated DNA transfection and the proton sponge hypothesis. J Gene Med 7, 657–663. 42. Kircheis, R., Wightman, L., and Wagner, E. (2001) Design and gene delivery activity of modified polyethylenimines. Adv Drug Deliv Rev 53, 341–358. 43. Brunner, S., Furtbauer, E., Sauer, T., Kursa, M., and Wagner, E. (2002) Overcoming the nuclear barrier: cell cycle independent nonviral gene transfer with linear polyethylenimine or electroporation. Mol Ther 5, 80–86. 44. Merdan, T., Kopecek, J., and Kissel, T. (2002) Prospects for cationic polymers in gene and oligonucleotide therapy against cancer. Adv Drug Deliv Rev 54, 715–758. 45. Neu, M., Fischer, D., and Kissel, T. (2005) Recent advances in rational gene transfer vector design based on poly(ethylene imine) and its derivatives. J Gene Med 7, 992–1009. 46. Oupicky, D., Ogris, M., and Seymour, L. W. (2002) Development of long-circulating polyelectrolyte complexes for systemic delivery of genes. J Drug Target 10, 93–98.

378

Sawant and Torchilin

47. Yamazaki, Y., Nango, M., Matsuura, M., Hasegawa, Y., Hasegawa, M., and Oku, N. (2000) Polycation liposomes, a novel nonviral gene transfer system, constructed from cetylated polyethylenimine. Gene Ther 7, 1148–1155. 48. Wang, D. A., Narang, A. S., Kotb, M., Gaber, A. O., Miller, D. D., Kim, S. W., and Mahato, R. I. (2002) Novel branched poly(ethylenimine)cholesterol water-soluble lipopolymers for gene delivery. Biomacromolecules 3, 1197–1207. 49. Lee, M., Rentz, J., Bikram, M., Han, S., Bull, D. A., and Kim, S. W. (2003) Hypoxiainducible VEGF gene delivery to ischemic myocardium using water-soluble lipopolymer. Gene Ther 10, 1535–1542. 50. Janat-Amsbury, M. M., Yockman, J. W., Lee, M., Kern, S., Furgeson, D. Y., Bikram, M., and Kim, S. W. (2005) Local, non-viral IL-12 gene therapy using a water soluble lipopolymer as carrier system combined with systemic paclitaxel for cancer treatment. J Control Release 101, 273–285. 51. Lee, M., Rentz, J., Han, S. O., Bull, D. A., and Kim, S. W. (2003) Water-soluble lipopolymer as an efficient carrier for gene delivery to myocardium. Gene Ther 10, 585–593. 52. Heyes, J., Palmer, L., Chan, K., Giesbrecht, C., Jeffs, L., and MacLachlan, I. (2007) Lipid encapsulation enables the effective systemic delivery of polyplex plasmid DNA. Mol Ther 15, 713–720. 53. Ko, Y. T., Kale, A., Hartner, W. C., Papahadjopoulos-Sternberg, B., and Torchilin, V. P. (2009) Self-assembling micelle-like nanoparticles based on phospholipid-polyethyleneimine conjugates for systemic gene delivery. J Control Release 133, 132–138. 54. Unger, E., Shen, D. K., Wu, G. L., and Fritz, T. (1991) Liposomes as MR contrast agents: pros and cons. Magn Reson Med 22, 304–308; discussion 313. 55. Barsky, D., Putz, B., Schulten, K., and Magin, R. L. (1992) Theory of paramagnetic contrast agents in liposome systems. Magn Reson Med 24, 1–13. 56. Unger, E., Tilcock, C., Ahkong, Q. F., and Fritz, T. (1990) Paramagnetic liposomes as magnetic resonance contrast agents. Invest Radiol 25 Suppl 1, S65–66. 57. Gries, H. (2002) Extracellular MRI contrast agents based on gadolinium., in Topics in current chemistry, contrast agent. (Krause, W., Ed.) pp 3-29, Springer-Verlag, Berlin Heidelberg. 58. Strijkers, G. J., Mulder, W. J., van Heeswijk, R. B., Frederik, P. M., Bomans, P., Magusin, P. C., and Nicolay, K. (2005) Relaxivity of lipo-

somal paramagnetic MRI contrast agents. Magma 18, 186–192. 59. Kabalka, G., Buonocore, E., Hubner, K., Moss, T., Norley, N., and Huang, L. (1987) Gadolinium-labeled liposomes: targeted MR contrast agents for the liver and spleen. Radiology 163, 255–258. 60. Kabalka, G. W., Davis, M. A., Moss, T. H., Buonocore, E., Hubner, K., Holmberg, E., Maruyama, K., and Huang, L. (1991) Gadolinium-labeled liposomes containing various amphiphilic Gd-DTPA derivatives: targeted MRI contrast enhancement agents for the liver. Magn Reson Med 19, 406–415. 61. Trubetskoy, V. S., Cannillo, J. A., Milshtein, A., Wolf, G. L., and Torchilin, V. P. (1995) Controlled delivery of Gd-containing liposomes to lymph nodes: surface modification may enhance MRI contrast properties. Magn Reson Imaging 13, 31–37. 62. McDannold, N., Fossheim, S. L., Rasmussen, H., Martin, H., Vykhodtseva, N., and Hynynen, K. (2004) Heat-activated liposomal MR contrast agent: initial in vivo results in rabbit liver and kidney. Radiology 230, 743–752. 63. Lokling, K. E., Fossheim, S. L., Skurtveit, R., Bjornerud, A., and Klaveness, J. (2001) pHsensitive paramagnetic liposomes as MRI contrast agents: in  vitro feasibility studies. Magn Reson Imaging 19, 731–738. 64. Trubetskoy, V. S., and Torchilin, V. P. (1994) New approaches in the chemical design of Gd-containing liposomes for use in magnetic resonance imaging of lymph nodes. J Liposome Res 4, 961–980. 65. Torchilin, V. P. (2000) Polymeric contrast agents for medical imaging. Curr Pharm Biotechnol 1, 183–215. 66. Weissig, V. V., Babich, J., and Torchilin, V. V. (2000) Long-circulating gadolinium-loaded liposomes: potential use for magnetic resonance imaging of the blood pool. Colloids Surf B Biointerfaces 18, 293–299. 67. Sims, G. E., and Snape, T. J. (1980) A method for the estimation of polyethylene glycol in plasma protein fractions. Anal Biochem 107, 60–63. 68. Karlsen, A., Blomhoff, R., and Gundersen, T. E. (2005) High-throughput analysis of vitamin C in human plasma with the use of HPLC with monolithic column and UV-detection. J Chromatogr B Analyt Technol Biomed Life Sci 824, 132–138. 69. Snyder, S. L., and Sobocinski, P. Z. (1975) An improved 2,4,6-trinitrobenzenesulfonic acid method for the determination of amines. Anal Biochem 64, 284–288.

Part IV Biofunctionalization of Surfaces and Thin Films

Chapter 24 Chemical Functionalization and Bioconjugation Strategies for Atomic Force Microscope Cantilevers Magnus Bergkvist and Nathaniel C. Cady Abstract Over the last decade, scanning probe microscopy (SPM) techniques, such as atomic force microscopy (AFM), have played an important role in a variety of biophysical research efforts. This straightforward technique has the capability to measure forces down to a few hundred piconewtons, which enables the observation of unique events within or between single molecules. However, in order to successfully carry out these types of biophysical measurements, the anchoring of the biomolecules of interest to the scanning probe cantilever tip needs to be of sufficient strength to avoid rupture prior to the analysis of the specific interaction to be probed. Hence, a covalent linkage of the biomolecule to the SPM probe tip is generally preferred. It is also advantageous to have a long-chain functional linker to separate the biomolecule from the SPM probe tip so as to minimize unwanted interactions between the substrate surface and the tip and to “isolate” the biomolecular forces being probed. The most common materials for SPM cantilevers are silica and silicon nitride, and there are several surface chemistry approaches available to achieve a covalent linkage to such types of materials. In this chapter, we present various strategies and detailed protocols for conducting surface modifications suitable for biomolecular attachment to AFM probe surfaces or other hydroxylated surfaces. The strategies described build upon an initial surface activation treatment using the convenient gas-phase deposition of an organosilane and incorporate various passivation schemes and biomolecular immobilization techniques. Key words: Scanning probe microscopy, Atomic force microscopy, Cantilever, AFM probe tip, Alkoxysilane, Aminosilane, Surface chemistry, PEGylation, Molecular vapor deposition

1. Introduction Efficient techniques for immobilizing biomolecules to interfaces are key in both industrial biotechnology as well as in fundamental life science research. One area of research that has gained much attention over the last 15 years is single-molecular biophysics, in large part due to the development of the atomic force microscope (AFM). AFM allows the study of the mechanical and physical properties of biomolecules, such as elasticity, stress response, Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_24, © Springer Science+Business Media, LLC 2011

381

382

Bergkvist and Cady

binding strength between receptor/target molecules, and more (1–5). For example, force spectroscopy can elucidate structural unfolding events in single proteins and can probe molecular interactions, such as antigen/antibody binding or DNA base pairing. Reliable force measurements require that the biomolecule of interest (e.g., protein, DNA, carbohydrate, etc.) is firmly attached to the AFM tip (or substrate) often via a linker molecule that reduces nonspecific interactions between biomolecules and the AFM tip. The linker can be a protein, such as streptavidin, with high affinity to a ligand (e.g., biotin) present on the biomolecule of interest. Other common linkers for biomolecule attachment to an AFM tip are linear poly(ethylene glycol) (PEG) chains incorporating various functional groups at the end (6–8). PEGs are chemically and physically inert and allow for free movement of the biomolecule when the AFM tip approaches the surface. Various strategies to covalently anchor proteins or functional PEG spacers to AFM tips can be devised, and a key component is the modification of the AFM tip to introduce a chemical “handle” for further conjugation. The derivatization of AFM tips with primary amines provides a versatile starting point for surface modification and a multitude of homo/heterofunctional reagents targeting amines are available for the covalent coupling of spacers/proteins. Organosilane reagents provide a convenient way to introduce reactive groups on surfaces with available hydroxyl groups, such as silicon oxide or silicon nitride AFM probes. Many protocols for silanization involve immersing a surface into a solution of an aminosilane reagent (in organic or aqueous solvents) for a period of time followed by rinsing and drying (9–11). This approach works for most applications and provides an adequate number of amine groups on the surface for bioconjugation. However, unless controlled environments and anhydrous solvents are used, several nanometer-thick layers of aminosilane can readily deposit due to the self-catalytic and crosslinking properties these reagents. Figure 1a shows the typical

Fig. 1.  AFM images of NH2-modified silica surfaces. (a) After immersion in an aminosilane/ toluene solution, followed by a toluene rinse; and (b) After molecular vapor deposition (MVD) of aminosilane (1 h, 60°C). Height scale: 0–25 nm.

Chemical Functionalization and Bioconjugation Strategies

383

appearance of a silica surface treated with an aminosilane (1 mM in toluene) for 2 h without working in a nitrogen (N2) glove box and not using anhydrous solvents. The AFM image shows the deposition of large aggregates and obvious surface roughening, both of which can affect the quality of biomolecule immobilization and impact the repeatability of nanoscale force measurements. Molecular vapor deposition (MVD) provides a gas-phase approach for the aminosilane derivatization of AFM tips (or other interfaces), which results in uniform molecular layers with reproducible quality. Figure 1b shows an example of an amine-modified surface produced using the MVD procedure. Gas-phase deposition is a proven method for the chemical modification of solid interfaces with various silanes (12, 13), including aminosilanes for AFM imaging of DNA and proteins (14, 15). In MVD, the silane reagent can be heated slightly to bring a low concentration of molecules into the vapor phase, which then uniformly deposit and covalently link to a solid interface without significant aggregation. In the case of aminosilanes, repeatable MVD can be realized between 60 and 90°C, which allows enough water to remain on the solid interface to promote the hydrolysis of the precursor and avoid significant multilayer formation. At the same time, the elevated temperature promotes the condensation of the aminosilane to give strong binding to the surface in a relatively short period of time. After the amination reaction, the AFM tip can be used for further conjugation with desired biomolecules or spacers. In this chapter, we describe a simple MVD approach to introduce a uniform and stable molecular film of reactive amine groups on silicon AFM tips and other oxide surfaces. The resulting amine functionality on the tip enables several routes for coupling biomolecules and/or spacer molecules, and we present here a few reliable and practical strategies to conjugate biomolecules to AFM tips either with or without a spacer functionality. An outline of some general strategies for surface modification is also provided.

2. Materials 2.1. Laboratory Equipment and Consumables

1. Silicon AFM probes (Veeco, mMash or other vendor). 2. Glass vials (Wheaton): Borosilicate glass scintillation vials with a Polyseal® cap liner, 20 ml; and borosilicate glass shell vials, 2 ml. 3. D-Salt™ dextran desalting columns, 5 ml (Pierce). 4. Glass microscope slides (Corning). 5. Polypropylene jar with screw cap lid, 32 oz (960  ml) (ThermoFisher).

384

Bergkvist and Cady

Fig.  2. Examples of flat-edge (top) and self-closing sharp-point (bottom) tweezers suitable for handling AFM probes.

6. Laboratory oven. 7. Rocking platform shaker. 8. Plasma cleaner equipped with a 4-in. quartz chamber and a vacuum pump (Harrick Plasma). 9. Disposable reagent reservoir, 25 ml, V-shaped, white polystyrene (Gilson). 10. Tweezers, with or without a Teflon coating (SPI supplies): Stainless steel flat-edge (“flat”) SPI-Swiss collection #2a; and self-closing stainless steel sharp-point (“sharp”) SPI-Swiss collection #N5 (Fig. 2). 2.2. Reagents and Other Chemicals

1. 3-Aminopropyltrimethoxysilane (APTMS) (Gelest). 2. Chloroform (CHCl3). 3. Dimethyl sulfoxide (DMSO). 4. Ethanol. 5. Ethanolamine. 6. Ethylenediaminetetraacetic acid, disodium salt (EDTA). 7. PEG compounds functionalized with various chemical groups. Examples of suggested reagents for conjugation include the following (Nanocs, Boston, MA): Thiol-PEGThiol, MW 5,000 (HS-PEG5000-HS); Thiol-PEG-Amine, MW 3,400 (HS-PEG3400-NH2); Thiol-PEG-Biotin, MW 3,400 (HS-PEG3400-Biotin); Thiol-PEG-Carboxyl, MW 3,400 (HS-PEG3400-COOH); NHS-PEG-Azido, MW 3,400 (NHS-PEG3400-N3); NHS-PEG-Maleimide, MW 3,400 (NHSPEG3400-Mal); and NHS-PEG-Biotin, MW 3,400 (NHSPEG3400-Biotin) (see Note 1).

Chemical Functionalization and Bioconjugation Strategies

385

8. Hydroxylamine. 9. Nitrogen gas (N2), minimum 99.999% purity (Airgas). 10. Sodium bicarbonate (NaHCO3). 11. Sodium cyanoborohydride (NaCNBH3). Caution: NaCNBH3 is toxic and must be used in a fume hood. Avoid acidic pH conditions when using this reagent. Store, handle, and dispose of all solutions containing NaCNBH3 according to appropriate safety procedures. 12. Sodium (meta)periodate (NaIO4). 13. Streptavidin. 14. Immunoglobulin G (IgG) antibody protein. 15. Succinimidyl 6-(3-[2-pyridyldithio]-propionamido)hexanoate (LC-SPDP) (Pierce). 16. N-Succinimidyl-S-acetylthioacetate (SATA) (Pierce). 17. Triethylamine (Et3N). 18. Deionized water (DI H2O), high-purity grade (18.2 MW cm resistivity). 2.3. Solutions

1. Solution A: 20  mM sodium phosphate, 0.15  M NaCl and 10 mM EDTA, pH 7.2. 2. Solution B: 0.5% (v/v) Et3N in chloroform, prepared by adding 10 ml of Et3N in 2 ml of chloroform. 3. Solution C: 0.1  M sodium phosphate, 0.15  M NaCl and 10 mM EDTA at pH 7.2. 4. Solution D: 0.5 M Hydroxylamine, 0.1 M sodium phosphate, 0.15 M NaCl, and 10 mM EDTA, pH 7.2. 5. Solution E: 20  mM sodium phosphate and 0.15  M NaCl, pH 7.2. 6. Solution F: 5 M NaCNBH3 in 1 M NaOH, prepared by dissolving 160  mg of NaCNBH3 in 0.5  ml of 1  M NaOH. Prepare 1 h before use and keep the solution inside a chemical fume hood at all times. 7. Solution G: 1 M Ethanolamine in solution E.

3. Methods In this section, we describe several versatile and effective techniques for the conjugation of linkers and biomolecules (proteins) to silicon AFM probe cantilever tips. Figure 3 outlines the various surface modification strategies, which are all based on the initial preactivation of an AFM tip with MVD-deposited aminosilane. Following amine surface activation, one route to biofunctionalize

386

Bergkvist and Cady O

Si

O Si

Si

A. Surface cleaning with air-plasma A. Plasma clean

B. Amino-silane treatment of surface using MVD C. Direct coupling of bifunctional NHS-PEG

HO

HO OH OH Si Si Si

C1. Protein conjugation to end-group functionality of PEG linker.

D. SPDP derivatization to obtain reactive pyridyl-disulfide

SPDP is smaller and less sterically hindered than the NHS-PEG in C. Coupling can be completed efficiently before hydrolysis of NHS. D1. Protein (such as avidin) with a thiol group can be immobilized on tip.

B. MVD of amine-silane

H2N

E. Coupling of bifunctional SH-PEG to pyridyl-disulfide surface

Advantage in that the thiol reaction is not sensitive to hydrolysis. Less PEG reagent is needed and possible to use longer reaction times. Thiol coupling allow alternative end-group chemistries and strategies for subsequent conjugation.

NH2

Si Si OH O O O OH O Si Si Si

D. SPDP/LC-SPDP activation

C. NHS-PEG-R2

S O

N

S

O

NH

E. HS-PEG-R 1

PEG

SH

protein with thiol group O NH

S

PEG

S

R2

NH

D1. Direct coupling to SPDP

R1

R1 = SH, COOH, NH 2, biotin O

E1. If R1 = NH2. Schiff’s base formation with aldehyde followed by reductive amination is one possibility. O

H

S

R2 = maleimide, biotin, N 3

C1. If R2 = maleimide

S

NH

NH

short link

O

O

PEG N

long link

O

S

protein with aldehyde group

O O

S

S NH

PEG NH

long link

Coupling chemistry is depending on R-group functionality. Possible strategies for E1 include: -COOH - EDC/NHS activation for amine coupling -SH - Disulfide exchange reaction with SPDP-activated protein -NH2 - Coupling to COOH groups in protein via carbodiimide linkage -NH2 - Schiff’s base formation with an aldehyde. -NH2 - Reaction with SPDP to introduce an active disulfide at end of PEG -NH2 - Reaction with NHS-maleimide reagent to couple proteins with a free thiol group.

Fig. 3. Outline of selected strategies for modifying AFM probe surfaces.

Chemical Functionalization and Bioconjugation Strategies

387

the AFM tip utilizes NHS-ester modified PEG linkers, whereas a second route uses active disulfide reagents for the immobilization of thiol-PEG linkers. The second route subsequently enables the coupling of biomolecules containing a free sulfhydryl (thiol) group to the AFM tip. The procedures presented below are grouped into sections that separately describe in detail the cleaning, amine-modification and coupling of functional linkers/biomolecules to AFM probe surfaces. 3.1. Cleaning of AFM Probes with an Air Plasma (see Fig. 3, Step A)

1. Use a pair of flat-edge (“flat”) tweezers to place an AFM probe on the surface of a glass microscope slide. To avoid damaging the tip of the AFM probe, orient it on the glass slide so that the pointed end of the cantilever tip faces up. 2. Place the glass slide with the AFM probe in the middle of the chamber of the plasma cleaner (Fig. 4). 3. Close the plasma cleaner door and start the vacuum pump to achieve 200–300  mTorr total negative pressure inside the chamber (see Note 2). 4. Once the vacuum stabilizes around 200–300  mTorr, adjust the power setting to “high” (~38 W) and turn on the plasma cleaner. Within 10–20 s, the plasma should ignite and a violet/ purple glow should be observed inside the chamber (due to N2). The length of time before the plasma ignites depends on the pressure inside the chamber and the power input (see Note 3).

Fig. 4. AFM probes sitting on a glass slide positioned inside the chamber of a plasma cleaner.

388

Bergkvist and Cady

5. Let the AFM probe sit in the ignited plasma for ~3  min (2–5 min is typically sufficient). This removes organic surface contaminants and activates the hydroxyl groups on the silica surface of the probe (see Notes 4 and 5). 6. Turn the plasma cleaner off, and stop the vacuum pump. Vent the chamber slowly by bleeding in air (or N2) into the chamber until atmospheric pressure is reached. Use the cleaned AFM probes as soon as possible, preferably within 10  min after venting the plasma chamber (see Note 6). Amine activation of AFM tips allows for linkage to amine-reactive cross-linkers or chemical groups on target molecules. Amines are reactive toward N-hydroxysuccinimidyl (NHS) esters, aldehydes, epoxides, etc., making them highly useful for a wide variety of chemical linking strategies. 1. Place a microscope slide carrying a cleaned AFM probe (obtained from Subheading 3.1) into a 32-oz (960-ml) polypropylene jar. 2. Add 1 ml of APTMS aminosilane to a 2 ml shell vial, and then spread out the reagent over the bottom of the vial with the pipette tip. Place the glass vial inside the polypropylene jar (which contains the glass slide with the AFM probe) such that the vial is positioned adjacent to the glass slide. 3. Seal the polypropylene jar using a screw cap lid and place it inside a 60°C oven for 60 min. 4. Remove the jar from the oven and let it cool to room temperature before removing the lid. The surface of the AFM cantilever tip (or any other substrate surface containing hydroxyl groups) now has a 1–2 molecule thick amine silane layer anchored to the surface. Ellipsometric thickness measurements performed on an APTMSmodified silicon wafer optical witness sample reveal the presence of a thin organic film of ~1 nm thickness, which fits well with the theoretical length (~0.9  nm) of the APTMS molecule (Fig. 5). 12

Thickness (Å)

3.2. Amine Activation of AFM Probe Surfaces (see Fig. 3, Step B)

10 8 6 4 2 0 Control

Amino-silane

Fig. 5.  Ellipsometric thickness measurements obtained for an unmodified silicon wafer and a silicon wafer modified with APTMS aminosilane using MVD. The control surface shows a slight increase in thickness that is likely due to the deposition of organic (carbon) contaminants from the ambient environment.

Chemical Functionalization and Bioconjugation Strategies

389

3500 3000 2500

N1s

a.u

MVD APTMS

2000

Control

1500 1000 500 0

406 404 402 400 398 396 394 392 390 Binding energy (eV)

Fig. 6. X-ray photoelectron spectroscopy (XPS) graph showing the N 1s intensity of the surface of an AFM probe modified with APTMS aminosilane via MVD as described in the text. The control AFM probe was incubated in a similar manner without any aminosilane present.

An X-ray photoelectron spectroscopy (XPS) graph of the N 1s intensity of the AFM cantilever base before and after silanization using the MVD technique is shown in Fig. 6, and provides further evidence (based on elemental composition) that the AFM probe has been chemically modified by APTMS. 3.3. Coupling of Functional NHS-Polyethylene Oxides to Aminated AFM Probe Surfaces (see Fig. 3, Step C)

Polyethylene oxides (PEOs), also frequently referred to as (PEGs), are commonly used as surface linkers to prevent nonspecific interactions with proteins or other biomolecules. This protocol uses succinimidyl ester (NHS) derivatives of PEG for direct linkage to amino-modified AFM tips. Although a 3.4-kDa PEG is used in this example protocol, a variety of other PEG or PEO molecules can also be used by following the same procedure. 1. Working in a chemical fume hood, dissolve 35  mg of NHSPEG3400-Mal in 2 ml of solution B (~5 mM PEG solution). Add the NHS-PEG solution to a clean, dry 20-ml glass vial resting on its side in a V-shaped reagent reservoir (see Note 7). 2. Use a pair of flat tweezers to gently immerse an amine-modified AFM probe (obtained from Subheading 3.2) into the NHSPEG solution with the tip facing upward and the cantilever oriented toward the opening of the glass vial (Fig. 7). 3. Carefully cap the glass vial and place the reagent reservoir carrying the vial on a rocking platform shaker. Place a glass microscope slide (~1 mm thick) under the reagent reservoir to tilt the vial, elevating the capped end slightly. 4. Incubate the AFM probe in the NHS-PEG solution for 2 h at room temperature while operating the rocking platform at the lowest agitation speed possible.

390

Bergkvist and Cady

Fig. 7. Illustration of an AFM tip inserted inside a glass vial resting on a V-shaped reagent tray (not to scale).

5. Prepare a second glass vial similarly to step 1, but containing 2 ml of pure chloroform for rinsing the AFM probe. 6. After the 2-h incubation period in step 4 is completed, carefully remove the AFM probe from the NHS-PEG solution with a pair of tweezers and transfer it into the rinse vial containing chloroform (prepared in step 5). Incubate the AFM probe in the chloroform solution for 1 min. 7. Use a pair of tweezers to temporarily remove the AFM probe from the rinse vial and replace the solution with 2 ml of fresh chloroform. Place the AFM probe back inside the vial and incubate further for 1 min. 8. Repeat step 7 once more so that the AFM probe has been rinsed a total of three times in chloroform. 9. After the final chloroform rinse, remove the AFM probe with a pair of tweezers and gently blow it dry with N2. 10. Preferably, use the PEG-modified AFM tip for further derivatization experiments immediately. Alternatively, the PEGmodified AFM probe can be stored for 1–2 days if kept dry in a desiccator and in the dark. 3.4. Introduction of Reactive Disulfides on Amine-Modified AFM Probe Surfaces Using LC-SPDP (see Fig. 3, Step D and Note 8)

Reactive disulfides can be used in a variety of different cross-linking strategies. Such disulfides react readily with naturally available or chemically/genetically introduced free sulfhydryl (thiol) groups on proteins and other biomolecules. This protocol describes the coupling of a heterobifunctional cross-linker containing a pyridyl disulfide group to amine-modified AFM probe surfaces via an NHS moiety. 1. Remove the LC-SPDP reagent from storage at −20°C and equilibrate it to room temperature in a desiccator before opening the container vial. 2. Working in a chemical fume hood, dissolve 4.25 mg of LCSPDP in 1 ml of DMSO to obtain a 10 mM LC-SPDP stock

Chemical Functionalization and Bioconjugation Strategies

391

Flat-edge tweezers holding chip.

AFM chip

Pointed, self-closing tweezers holding across chip body.

Fig. 8. Illustration of how the AFM probe should be handled with pairs of flat-edge tweezers and self-enclosing sharp-point tweezers while being transferred between solution reservoirs.

solution. 3. Add 200 ml of the LC-SPDP stock solution to 2.0 ml of solution A, and then transfer this mixture to a clean 20-ml glass vial resting on its side in a V-shaped reagent reservoir. 4. Use a pair of flat tweezers to gently immerse an aminemodified AFM probe (obtained from Subheading 3.2) into the LC-SPDP solution with the tip facing upward and the cantilever oriented toward the opening of the glass vial (Fig. 7). 5. Carefully cap the glass vial and place the reagent reservoir carrying the vial on a rocking platform shaker. Place a glass microscope slide (~1 mm thick) under the reservoir to tilt the vial, elevating the capped end slightly. 6. Incubate the AFM probe in the LC-SPDP solution for 45 min at room temperature while operating the rocking platform at the lowest agitation speed possible. 7. After the 45-min incubation period is completed, carefully remove the AFM probe from the LC-SPDP solution with a pair of flat tweezers. While holding the base of the AFM probe with the flat tweezers, use a pair of self-closing sharppoint (“sharp”) tweezers to grip the probe across the silicon chip-body (Fig. 8; see Notes 9 and 10). 8. While holding the AFM probe with the sharp tweezers, use a squirt bottle containing DI H2O to carefully rinse the cantilever, applying water to the backside of the chip base and letting it flow forward down toward and over the cantilever tip (see Note 11). 9. Dry the tip with a stream of N2 gas while still gripping the AFM probe with the self-closing sharp tweezers. After completing the above procedures, the presence of thiols on the AFM probe can be verified by using an XPS (Fig.  9). The cantilever/tip of the AFM probe is now activated with reactive pyridyl disulfide groups that can be used for further conjugation reactions with proteins or linkers, such as PEGs (see Note 12).

392

Bergkvist and Cady 3500 3000 2500

S2p MVD Amine-silane + LC-SPDP MVD Amine control

a.u.

2000 1500 1000 500 0 172

170

168 166 164 162 Binding Energy (eV)

160

Fig. 9. XPS graph showing the S 2p intensity of the surface of an amine-activated AFM probe after reaction with LC-SPDP as described in the text. The control AFM probe was incubated in a similar manner without any LC-SPDP present.

3.5. Coupling of Thiol-Polyethylene Oxides to DisulfideActivated AFM Probe Surfaces (see Fig. 3, Step E and Note 13)

Reactive disulfides present on an AFM tip allow linking of thiol-modified compounds, such as thiolated PEG (HS-PEG). This provides yet another route to carry out the passivation of surfaces or the attachment of high-molecular weight spacer molecules. In general, thiol chemistry is less sensitive to hydrolysis in aqueous solution compared to NHS chemistry, and thus enables the use of longer incubation times for efficient conjugation. 1. Dissolve 10  mg of HS-PEG5000-SH in 2  ml of solution A (1  mM PEG solution), and add the resulting HS-PEG-SH solution to a clean 20-ml glass vial resting on its side in a V-shaped reagent reservoir (see Note 14). 2. Use a pair of flat tweezers to gently immerse an SPDPactivated AFM tip into the HS-PEG-SH solution with the tip facing upward and the cantilever oriented toward the opening of the glass vial (Fig. 7). 3. Carefully cap the glass vial and place the reagent reservoir carrying the vial on a rocking platform shaker. Place a glass microscope slide (~1 mm thick) under the reservoir to tilt the vial, elevating the capped end slightly. 4. Incubate the AFM probe in the HS-PEG-SH solution for 2 h at room temperature while operating the rocking platform at the lowest agitation speed possible. 5. After the 2-h incubation period, carefully remove the AFM probe from the solution with a pair of flat tweezers. While holding the base of the AFM probe with the flat tweezers, use a pair of self-closing sharp tweezers to grip the AFM probe across the silicon chip-body (Fig. 8).

Chemical Functionalization and Bioconjugation Strategies

393

6. Immerse the AFM probe into a clean beaker containing 25 ml of solution A, and gently move the probe back and forth for ~1  min to remove excess HS-PEG-SH reagent. Proceed to step 7 without drying the tip in-between. 7. The AFM cantilever tip now contains a PEG linker with a free thiol end-group. To avoid unwanted oxidation of the PEGthiol groups to the disulfide form, immediately couple a biomolecule or other conjugation agent containing a thiolreactive group (e.g., maleimide or pyridyl disulfide) to the AFM tip. AFM probes modified with PEG compounds containing reactive end groups other than the sulfhydryl group can be rinsed in DI-H2O, dried under N2 and stored for 1–2 days if kept dry in a vacuum desiccator and in the dark. 3.6. Preparation of Thiolated Streptavidin (SA-SH) Using SATA

Proteins with free thiols can be directly linked to AFM tips modified with maleimide or activated disulfide groups. This protocol describes a method to introduce free thiol groups into streptavidin, which can subsequently be used for conjugation to an AFM tip.

3.6.1. Reaction of Streptavidin with SATA

1. Dissolve streptavidin (SA) in solution C to obtain a 5 mg/ml SA solution. 2. Working in a chemical fume hood, dissolve SATA in DMSO to obtain a 12 mg/ml (~52 mM) SATA solution. 3. Add 12.5 ml of the SATA solution to each ml of SA solution to be modified. 4. Incubate the reaction mixture for 30  min at room temperature. 5. Purify the SATA-SA product using a D-Salt™ desalting ­column equilibrated with solution C according to the manufacturer’s instructions. Alternatively, dialyze the sample against solution C for 24 h using a dialysis membrane (3.5 kDa MWCO). The purified SATA-SA product can be stored for an extended period of time at 4°C.

3.6.2. Deprotection of SATA-Modified Streptavidin to Generate Free Sulfhydryl Groups

To generate free sulfhydryl groups suitable for use in AFM-tip conjugation reactions, the SATA-modified streptavidin (SATA-SA) (obtained from Subheading 3.6.1) is deacetylated using the following procedure. 1. Add 25 ml of solution D to 225 ml of purified SATA-SA. 2. Incubate the reaction mixture for 2 h at room temperature to generate SA containing free thiol groups (SA-SH).

3.6.3. Purification of Sulfhydryl-Modified Streptavidin (SA-SH)

Purification of SA-SH is performed using a D-Salt™ desalting column (5 ml) as described below.

394

Bergkvist and Cady

1. Equilibrate a D-Salt™ column with 25 ml of solution A. 2. Add the reaction mixture (obtained from Subheading 3.6.2, step 2) containing sulfhydryl-modified streptavidin to the D-Salt™ column and let the solution enter the column completely. The sample volume (250 ml) loaded corresponds to 5% of the total column volume. 3. Add 1.5 ml of solution A to the column and discard the eluting fraction. 4. Add an additional 1.5 ml of solution A in 250-ml portions and collect the individual fractions that elute from the column. 5. Monitor the presence or absence of protein in each fraction via UV absorbance measurements taken at 280  nm wavelength. Most of the SA-SH protein should elute within the first four collected fractions. Pool the fractions containing purified SA-SH protein. 6. Use the purified SA-SH protein immediately for further conjugation experiments. 3.7. Coupling of Thiolated Streptavidin (SA-SH) to AFM Tips Modified with a Thiol-Reactive Group (see Fig. 2, Steps C1 and D1)

Proteins containing free thiol groups can be directly linked to AFM tips modified with sulfhydryl-reactive groups. In the example protocol provided below, we describe the coupling of thiolated-streptavidin to an AFM probe surface modified with either a maleimide group or a pyridyl disulfide group. Such types of thiol-reactive groups can be introduced onto the AFM tip using a short linker (e.g., LC-SPDP, as described in Subheading 3.4), or a longer flexible PEG-linker (as described in Subheading 3.3). 1. Add 1.0  ml of solution A to 1.0  ml of purified SA-SH (obtained from Subheading 3.6.3), and transfer the mixture to a 20-ml glass vial resting on its side in a V-shaped reagent reservoir. The final concentration of SA-SH (or any other desired protein to be conjugated) in solution A should be between 0.25 and 1.0 mg/ml of protein. 2. Use a pair of flat tweezers to gently immerse an AFM probe modified with a thiol-reactive group (e.g., maleimide or pyridyl disulfide) into the SA-SH solution with the tip facing upward and the cantilever oriented toward the opening of the glass vial (Fig. 7). 3. Carefully cap the vial and place the reagent reservoir carrying the vial on a rocking platform shaker. Place a microscope slide (~1 mm thick) under the reservoir to tilt the vial, elevating the capped end slightly. 4. Incubate the AFM probe in the SA-SH solution for 2  h at room temperature while operating the rocking platform at the lowest speed possible. If necessary, coupling can also be performed at 4°C for 16 h.

Chemical Functionalization and Bioconjugation Strategies

395

5. After the incubation period, carefully remove the AFM probe from the solution with a pair of flat tweezers. While holding the base of the AFM probe with the flat tweezers, use a pair of self-closing sharp tweezers to grip the AFM probe across the silicon chip-body (Fig. 8). 6. Immerse the AFM probe into a clean beaker containing 25 ml of solution A, and gently move the probe back and forth for ~1 min to remove excess SA-SH protein. Dry the tip with a stream of N2 gas while still gripping the AFM probe with the self-closing sharp tweezers. 7. Use the SA-conjugated probe tip immediately for AFM experiments. 3.8. Activation of Glycoproteins with Sodium (Meta) Periodate

3.8.1. Reaction of IgG Glycoproteins with Sodium (Meta)Periodate

In the example protocol described below, the carbohydrate groups present on IgG are oxidized with sodium (meta)periodate to generate reactive aldehydes for subsequent conjugation to AFM probes functionalized with an amine-modified PEG-NH2 compound. Other types of glycoproteins can also be oxidized similarly using these procedures. 1. Dissolve IgG in solution E to obtain a 5–10 mg/ml solution. 2. Dissolve NaIO4 in DI H2O to obtain a 0.1 M solution. Protect the NaIO4 solution from light. 3. Add 100 ml of the NaIO4 solution to each ml of the IgG solution and mix well. Protect the reaction mixture from light. 4. Incubate the reaction mixture at room temperature for 30 min in the dark.

3.8.2. Purification of Oxidized IgG Glycoproteins

Immediately purify the oxidized IgG product prepared in Subheading 3.8.1 using a D-Salt™ desalting column as follows: 1. Equilibrate a D-Salt™ column with 25 ml of solution E. 2. Load 250  ml of the oxidized IgG sample (obtained from Subheading  3.8.1, step 4) onto the D-Salt™ column. If another type of desalting column is used instead, do not load more than 5% of the total column volume in order to achieve efficient separation between the oxidized IgG product and excess NaIO4. 3. Add 250-ml aliquots of solution E to the column and collect the eluted fractions (250 ml) while measuring the UV absorbance at 280 nm wavelength to monitor for protein content. The oxidized IgG product should start to elute from the column after a total of 1.75 ml of solution E has been added (including the sample volume). The majority of protein should elute from the D-salt™ column over the next 1-ml volume of solution E added (i.e. between 1.75–2.75 ml total added volume).

396

Bergkvist and Cady

4. Pool the eluted fractions containing the purified protein product. Immediately use the oxidized IgG protein for coupling to PEG-NH2-modified AFM probes. 3.9. Coupling of Glycoproteins to SH-PEG-NH2-Modified AFM Tips via Reductive Amination (see Fig. 3, Step E1)

This protocol assumes that the glycoprotein of interest has been oxidized to contain reactive aldehyde groups (see Subheading 3.8). 1. Add 1.0 ml of solution E to 1.0 ml of glycoprotein, and transfer the solution to a 20-ml glass vial resting on its side in a V-shaped reagent reservoir. The final protein concentration should be at a minimum of 0.5 mg/ml. Adjust the relative volumes of solution E and glycoprotein if needed. 2. Use a pair of flat tweezers to gently immerse an AFM probe modified with a PEG-NH2 group into the glycoprotein solution with the tip facing upward and the cantilever oriented toward the opening of the glass vial (Fig. 7). 3. Carefully cap the glass vial and place the reagent reservoir carrying the vial on a rocking platform shaker. Place a glass microscope slide (~1 mm thick) under the reservoir to tilt the vial, elevating the capped end slightly. 4. Incubate the AFM probe in the glycoprotein solution for 1.5 h at room temperature while operating the rocking platform at the lowest agitation speed possible (see Note 15). 5. After the 1.5-h incubation period, carefully remove the AFM probe from the solution with a pair of flat tweezers. While holding the base of the AFM probe with the flat tweezers, use a pair of self-closing sharp tweezers to grip the AFM probe across the silicon chip body (Fig. 8). 6. Immerse the AFM probe into a clean beaker containing 25 ml of solution E and gently move the probe back and forth for ~1 min to remove excess glycoproteins. 7. Place the AFM probe into a clean glass vial containing 2.0 ml of solution E, resting on its side in a V-shaped reagent reservoir in a fume hood. (Use a new glass vial for this step.) 8. While working in a fume hood, add 20 ml of Solution F to the glass vial (see Note 16). (Solution F should be prepared 1 h before use, and always kept inside a fume hood). 9. Carefully cap the glass vial and place the reagent reservoir carrying the vial on a rocking platform shaker inside the fume hood. Place a glass microscope slide (~1 mm thick) under the reservoir to tilt the vial, elevating the capped end slightly. 10. Incubate the AFM probe inside the glass vial for 30 min at room temperature while operating the rocking platform at the lowest agitation speed possible. 11. Add 100 ml of Solution G to the glass vial (see Note 17).

Chemical Functionalization and Bioconjugation Strategies

397

12. Carefully cap the glass vial and incubate the AFM probe for an additional 30 min at room temperature, again using the rocking platform shaker inside the fume hood. 13. After the 30-min incubation period, carefully remove the AFM probe from the solution with a pair of flat tweezers. While holding the base of the AFM probe with the flat tweezers, use a pair of self-closing sharp tweezers to grip the AFM probe across the silicon chip-body (Fig. 8). 14. Immerse the AFM probe into a clean beaker containing 25 ml of solution E and gently move the probe back and forth for ~1 min to remove excess reagents (see Note 18). 15. Use glycoprotein-conjugated probe tip immediately for AFM experiments.

4. Notes 1. The PEG linkers listed in Subheading 2.2 are some examples of commercially available reagents that feature a relatively long spacer length (5–10  nm). Depending on the specific application, the use of shorter PEG linkers might be more suitable; the protocols described in Subheading 3 can also be used in conjunction with such types of shorter linkers. 2. The described plasma cleaner system is typically equipped with vacuum pumps that pump down to an ultimate negative total pressure of 200  mTorr or lower (“negative pressure” refers to a pressure reading below atmospheric pressure). If this is the case, open the vent valve on the plasma door slightly to bleed in air to stabilize the total pressure inside the chamber at negative 200–300 mTorr. 3. If the vacuum is not low enough, there will be “too many molecules” in the chamber to ignite the plasma. On the other hand, if the vacuum is too low there will not be enough molecules to sustain the plasma, at which point the intensity will drop and eventually disappear, i.e., the plasma “dies.” 4. In performing these procedures, we have observed that about 1–2  nm of organic contaminants are removed by a 2-min plasma exposure under the described conditions. Plasma cleaning is thus a very mild cleaning treatment to remove very thin layers of carbon, and the use of longer exposure times/higher power may be needed if the sample is heavily contaminated. Typically, new AFM probes received from commercial vendors have only a thin layer of organic contaminants and so the cleaning protocol described here should be sufficient.

398

Bergkvist and Cady

5. In addition to removing hydrocarbon (organic) contaminant material, the plasma also helps break surface bonds to generate hydroxyl groups for chemical coupling. 6. The described procedures work just as well with a plasma cleaner equipped with a 2-in. diameter quartz chamber. Other instrument setups using plasma (Air, O2) can also be used for this initial cleaning step; however, the exposure times, power settings and vacuum levels may need to be adjusted to achieve efficient surface decontamination. 7. Any of the other NHS-PEGs listed in Subheading  2.2 can also be used in a similar fashion. 8. Alternatively, SPDP, which contains a shorter carbon spacer than LC-SPDP, can also be used here as well. The length of the carbon spacer arm for LC-SPDP/SPDP is 15.7 Å/6.8 Å, respectively. 9. If the self-closing sharp-point tweezers become bent or crooked, you can realign/sharpen them using diamond lapping polishing paper (commonly used for preparing TEM samples for cross-section analyses) with between 0.1 and 3  mm particle (grit) size. Simply fold the polishing paper around the base of the sharp tip and pull the tweezers back in a uniform motion. Repeat until the tips of the tweezers become sharp and straight. Keep in mind, however, that this eventual destroys any Teflon coating on the tweezers. 10. The AFM probe can be placed onto a clean-room tissue wipe before being picked up with a pair of self-closing sharp tweezers if attempting to manipulate two tweezers at once is problematic. We prefer using the “two tweezers” approach to avoid placing the AFM probe on a solid surface before performing the rinsing/drying steps. 11. Alternatively, the AFM probe can be rinsed by dipping it into a clean beaker containing 25  ml of DI H2O with a pair of tweezers, and gently moving the probe back and forth for 1 min inside the beaker. Holding the AFM probe with a pair of tweezers (with the cantilever tip facing downward) under a stream of water coming directly from the outlet tap of a DI H2O purification system also works well if the flow rate is not too high. 12. Introducing a reactive disulfide on the AFM tip allows conjugation to a protein containing a free thiol. If the protein does not have a free thiol group in its native structure, then free thiol groups can be introduced by treating the protein with an amine-reactive thiolation reagent, such as Traut’s reagent (2-iminothiolane⋅HCl). If a coupling reagent containing a disulfide group (such as SPDP) is used instead, the disulfide on the protein will need to be reduced in order to generate a free thiol. These reduction reactions are commonly performed

Chemical Functionalization and Bioconjugation Strategies

399

with dithiothreitol (DTT) or other reducing agents (e.g., tris(2-carboxyethyl)phosphine hydrochloride, TCEP). After reduction, the protein is purified over a desalting column with EDTA added to the eluting buffer solution in order to prevent reoxidation. The reverse strategy can also be envisioned, where the active disulfide resides on the protein and the SPDP coupled to the AFM tip is reduced to give a free sulfhydryl group. However, the close proximity between the resultant free thiols on the surface of the AFM tip can lead to rapid reoxidation and the formation of surface disulfides that are relatively inactive. Consequently, this can lower the conjugation efficiency with the biomolecule containing the reactive disulfide. 13. Any of the other thiol-PEGs listed in Subheading 2.2 can also be used in a similar fashion. 14. Thiol-containing PEG reagents can spontaneously oxidize to form disulfide-PEGs. If this happens, their reactivity toward moieties such as SPDP and maleimides is significantly reduced. If disulfide formation is suspected, the thiol-PEG can be prepared in solution A containing 25  mM DTT, which will reduce any disulfides within a few minutes. The solution is then passed over a desalting column (in a similar fashion as described for SA-SH in Subheading 3.6.3) to obtain SH-PEG. If dilution of the thiol-PEG reagent is a concern, then use a reducing agent immobilized to a solid matrix, such as immobilized TCEP, which can be removed without diluting the initial sample preparation. 15. This step generates a Schiff’s base linkage between PEG-NH2 and the aldehyde group present on the glycoprotein. 16. This step reduces the relatively labile Schiff base linkage to a highly stable amine bond. 17. This step blocks any remaining aldehyde groups present on the glycoprotein. 18. Other buffers can also be used for rinsing as long as they do not contain any thiol-reducing agents. References 1. Forbes, J. G.; Jin, A. J.; and Wang, K., (2001) Atomic force microscope study of the effect of the immobilization substrate on the structure and force extension curves of a multimeric protein. Langmuir. 17, 3067–3075. 2. Puchner, E. M.; Alexandrovich, A.; Kho, A. L.; Hensen, U.; Schäfer, L. V.; Brandmeier, B.; Gräter, F.; Grubmüller, H.; Gaub, H. E.; and Gautel, M. (2008) Mechanoenzymatics of titin kinase. Proc. Natl Acad. Sci. USA 105, 13385–13390.

3. Yu, J.; Sun, S.; Jiang, Y.; Ma, X.; Chen, F.; Zhang, G.; and Fang, X. (2006) Single molecule study of binding force between transcription factor TINY and its DNA responsive element. Polymer. 47, 2533–2538. 4. Ros, R.; Schwesinger, F.; Anselmetti, D.; Kubon, M.; Schafer, R.; Pluckthun, A.; and Tiefenauer, L. (1998) Antigen binding forces of individually addressed single-chain fv antibody molecules. Proc. Natl Acad. Sci. USA 95, 7402–7405.

400

Bergkvist and Cady

5. Carrión-Vázquez, M.; Oberhauser, A.; Díez, H.; Hervás, R.; Oroz, J.; Fernández, J.; and Martínez-Martín, D. (2006) Protein nanomechanics – as studied by AFM single-molecule force spectroscopy. In Advanced Techniques in Biophysics (Arrondo, J.L.R and Alonso, A., eds.), Springer, New York, NY. pp 163–245. 6. Riener, C. K.; Kienberger, F.; Hahn, C. D.; Buchinger, G. M.; Egwim, I. O. C.; Haselgrübler, T.; Ebner, A.; Romanin, C.; Klampfl, C.; Lackner, B.; Prinz, H.; Blaas, D.; Hinterdorfer, P.; and Gruber, H. J. (2003) Heterobifunctional crosslinkers for tethering single ligand molecules to scanning probes. Anal. Chim. Acta. 497, 101–114. 7. Chen, G.; Ning, X.; Park, B.; Boons, G.-J.; and Xu, B. (2009). Simple, clickable protocol for atomic force microscopy tip modification and its application for trace ricin detection by recognition imaging. Langmuir. 25, 2860–2864. 8. Lee, H.; Scherer, N. F.; and Messersmith, P. B. (2006) Single-molecule mechanics of mussel adhesion. Proc. Natl. Acad. Sci. USA 103, 12999–13003. 9. Etienne, M. and Walcarius, A. (2003) Analytical investigation of the chemical reactivity and stability of aminopropyl-grafted silica in aqueous medium. Talanta. 59, 1173–1188. 10. Moon, J. H.; Shin, J. W.; Kim, S. Y.; and Park, J. W. (1996) Formation of uniform aminosilane thin layers: An imine formation to

­ easure relative surface density of the amine m group. Langmuir. 12, 4621–4624. 11. Flink, S.; van Veggel, F.; and Reinhoudt, D. N. (2001) Functionalization of self-assembled monolayers on glass and oxidized silicon wafers by surface reactions. J. Phys. Org. Chem. 14, 407–415. 12. Ashurst, R. W.; Carraro, C.; Chinn, J. D.; Fuentes, V.; Kobrin, B.; Maboudian, R.; Nowak, R.; and Yi, R. (2004) Improved vaporphase deposition technique for antistiction monolayers. In Micromachining and Microfabrication Process Technology, Volume 5342 of the Proceedings of SPIE, San Jose, CA, pp. 204–211. 13. Finocchio, E.; Macis, E.; Raiteri, R.; and Busca, G. (2007) Adsorption of trimethoxysilane and of 3-mercaptopropyltrimethoxysilane on silica and on silicon wafers from vapor phase: An IR study. Langmuir. 23, 2505–2509. 14. Lyubchenko, Y.; Shlyakhtenko, L.; Harrington, R.; Oden, P.; and Lindsay, S. (1993) Atomic force microscopy of long DNA - imaging in air and under water. Proc. Natl Acad. Sci. USA 90, 2137–2140. 15. Bergkvist, M.; Carlsson, J.; Karlsson, T.; and Oscarsson, S. (1998) TM-AFM Threshold analysis of macromolecular orientation: a study of the orientation of IgG and IgE on mica surfaces. J. Colloid Interface Sci. 206, 475–481.

Chapter 25 Chemoselective Protein and Peptide Immobilization on Biosensor Surfaces Edith H.M. Lempens, Brett A. Helms, and Maarten Merkx Abstract Site-specific immobilization of proteins and peptides on a sensor surface represents a significant challenge for bioanalytical applications such as surface plasmon resonance (SPR). The most common protocols for covalent protein immobilization usually result in heterogeneous presentation of the ligand at the surface, which can in some instances yield conflicting results with analogous data obtained in solution. Here, we discuss two complementary and generic bioconjugation methods that allow chemoselective immobilization of peptides and proteins via either their C-terminus (native chemical ligation) or their N-terminus (oxime ligation). While the protocols described in this chapter were designed for use in a Biacore instrument, the methods should also be applicable to other SPR instruments and, with slight adjustments, to many other types of bioanalytical applications that rely on protein-functionalized surfaces. Key words: Biosensor, Protein immobilization, Surface plasmon resonance, Biacore, Native chemical ligation, Oxime ligation

1. Introduction The immobilization of proteins on surfaces is of practical importance to many areas of the life sciences. The combination of surface plasmon resonance (SPR) with a microfluidic delivery system such as that used in, for example, Biacore instruments is a powerful technology for obtaining kinetic as well as thermodynamic data on biomolecular interactions. However, the site-specific immobilization of proteins and peptides on a sensor surface represents a significant challenge for SPR. The most common protocols for covalent protein immobilization target amine or thiol groups at the protein exterior; consequently, the use of these methods typically results in heterogeneous presentation of the protein at the surface, which can in some cases yield conflicting Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_25, © Springer Science+Business Media, LLC 2011

401

402

Lempens, Helms, and Merkx

results with analogous data obtained in solution (1). To date, several approaches have been developed for Biacore instruments that allow for more controlled and homogeneous ligand presentation on the chip surface, including the use of biotinylated ligands on streptavidin-functionalized chips, Ni-NTA chips for the immobilization of His-tagged proteins, and the use of antibodies to present the protein ligand in a single orientation. While these methods have proven to be valuable alternatives to direct coupling approaches in certain cases, they all have their own specific drawbacks (2). In recent years, we as well as others have developed several site-specific covalent immobilization strategies based on the use of bio-orthogonal ligation reactions that were originally developed in the field of peptide chemistry (3). In particular, we explored two complementary and generic bioconjugation methods that allow for the chemoselective immobilization of peptides or recombinant proteins to surfaces via either their C-terminus (native chemical ligation) (4) or their N-terminus (oxime ligation) (5) (Fig. 1). In general, bioconjugation methods targeting the flexible termini of proteins are least likely to interfere with their function. Another attractive property of both conjugation

Fig. 1. Classical methods for the immobilization of proteins via reaction with their surface-accessible amines (a) result in heterogeneous presentation of the proteins at surfaces. Conjugation using native chemical ligation (b) or oxime ligation (c) allows homogenous presentation of proteins through chemoselective immobilization via the protein’s C- or N-terminus, respectively.

Chemoselective Protein and Peptide Immobilization on Biosensor Surfaces

403

methods is that the surface density of the immobilized ligands is highly controllable and can be increased stepwise at any point in the sensor chip’s lifetime. In this chapter, we present detailed protocols for the preparation of SPR chips for use in either native chemical ligation or oxime ligation, the preparation of N- and C-terminally reactive proteins and peptides, and examples of typical applications. While the protocols described herein were originally designed for use in a Biacore instrument, the methods should also be applicable to other SPR instruments and, with slight adjustments, to many other types of bioanalytical applications that rely on protein-functionalized surfaces.

2. Materials 2.1. Synthesis of Thiazolidine-Protected Cysteine Linker (4)

1. N-(tert-Butyloxycarbonyl)thiazolidine carboxylic acid. 2. N-Hydroxysuccinimide (NHS). 3. 2-(1H-benzotriazole-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate (HBTU). 4. N,N-Diisopropylethylamine (DIPEA). 5. N,N-dimethylformamide (DMF). 6. Diethyl ether. 7. Saturated KCl. 8. Saturated NaHCO3. 9. Na2SO4. 10. Mono-N-(tert-butoxycarbonyl)ethylenediamine was obtained by protecting one amino group of ethylenediamine with tertbutyl phenylcarbonate, as described previously in ref. 6. 11. Dichloromethane (DCM). 12. Trifluoroacetic acid (TFA). 13. Ethyl acetate. 14. Brine (saturated NaCl).

2.2. Functionalization of SPR Chips with N-Terminal Cysteines

1. Biacore T100 SPR system (GE Healthcare). 2. Biacore CM5 sensor chips (GE Healthcare). 3. HBS-EP buffer (10  mM HEPES, 150  mM NaCl, 3  mM EDTA, 0.5% (v/v) surfactant P20, pH 7.4) is obtained as a 10× concentrated stock solution from GE Healthcare. After appropriate dilution, the buffer is filtered through 0.2-mm nylon ZapCap-CR filters (Whatman). 4. 1.0 M solution of ethyl-3-[3-dimethylaminopropyl]carbodiimide HCl (EDC) in deionized water.

404

Lempens, Helms, and Merkx

5. 1.0 M solution of N-hydroxysuccinimide (NHS) in deionized water. 6. 250  mM compound 4 (Fig.  3) in 50  mM borate buffer, pH 8.5. 7. Ethanolamine HCl. 8. 250 mM methoxyamine HCl in 50 mM acetate buffer, pH 4.0. 2.3. Synthesis of Peptides with a C-Terminal b-Mercaptopro­pionic Acid-Leucine Thioester

1. All tert-butyloxycarbonyl (t-Boc)-protected amino acids were obtained from Novabiochem. 2. 4-Methylbenzhydrylamine hydrochloride (MBHA) resin (0.92 mmol/g loading) (Novabiochem). 3. S-Trityl 3-mercaptopropionic acid (Sigma). 4. Triisopropylsilane (Aldrich). 5. HF containing 4% (v/v) p-cresol (see Note 1). 6. Reversed-phase high-performance liquid chromatography (HPLC) was performed on a Varian Prostar 320 HPLC system equipped with a VYDAC® protein/peptide C18 column. A gradient of acetonitrile (ACN) in water, both containing 0.1% (v/v) TFA, was used to elute the peptides.

2.4. Synthesis of Recombinant Green Fluorescent Protein with a C-Terminal 2-Mercaptoethane­ sulfonic Acid Thioester

1. Plasmid pGFPX: pTXB1 vector [New England Biolabs (NEB)] with a green fluorescent protein (GFP) gene cloned into the NdeI and EcoRI sites present in the multiple cloning site (MCS). This vector encodes GFP fused at its C-terminus to an Mxe intein and a chitin binding domain (7). Other cloning vectors using different intein sequences are also available from NEB. 2. Competent Escherichia coli BL21(DE3) cells (Novagen). 3. LB medium (10  g Bacto-peptone, 5  g yeast extract, 10  g NaCl per liter of solution) containing 100  mg/L of ampicillin. 4. Isopropyl b-d-1-thiogalactopyranoside (IPTG). 5. Shaking incubator. 6. High-speed centrifuge. 7. BugBuster® (Novagen) cell lysis buffer containing 1 mL/mL Benzonase® Nuclease (Novagen). 8. 10-mL chitin column. 9. Column buffer: 20  mM sodium phosphate, 0.5  M NaCl, 0.1 mM EDTA, pH 8.0. 10. Cleavage buffer: 200 mM sodium phosphate, 0.5 M NaCl, 0.1  mM EDTA, 50  mM 2-mercaptoethanesulfonic acid (MESNA), pH 6.0.

Chemoselective Protein and Peptide Immobilization on Biosensor Surfaces

405

2.5. Peptide/Protein Ligation to CysteineFunctionalized SPR Chips

1. SPR chip with N-terminal cysteines (see Subheading 2.2).

2.6. Synthesis of t-Boc-Protected Aminooxy Linker (8)

1. Benzyl phenyl carbonate was synthesized according to previously described literature procedures (8).

2. Peptides or proteins with a C-terminal thioester (see Subheading 2.3 or 2.4). 3. HBS-EP buffer (pH 7.4) containing 50 mM 4-mercaptophenylacetic acid (MPAA) and 10  mM tris(carboxyethyl)phosphine HCl (TCEP).

2. Ethylenediamine. 3. Ethanol. 4. 3 M HCl. 5. 5 M NaOH. 6. N-[(tert-butyloxycarbonyl)-aminooxy]acetic acid. 7. 2-(6-Chloro-1H-benzotriazole-1-yl)-1,1,3,3-tetramethyl­ ammonium hexafluorophosphate (HCTU). 8. Palladium (10%, w/w) on activated carbon (Pd/C catalyst). 9. Hydrogen gas. 10. Celite® 545, diatomaceous earth.

2.7. Functionalization of SPR Chips with Aminooxy Groups

The same materials as listed in Subheading 2.2 are used (with the exception of items 6 and 8). 1. 250  mM compound 8 (Fig.  6) in 50  mM borate buffer, pH 8.5. 2. 1.0 M Phosphoric acid, pH 2.0.

2.8. Synthesis of Peptides or Proteins with an N-Terminal Aldehyde Group 2.9. Synthesis of Peptides or Proteins with an N-Terminal Ketone Group

The peptides can be prepared by using either Fmoc- or t-Bocmediated solid-phase peptide synthesis. 1. NaIO4. 2. 10 mM Sodium phosphate buffer, pH 7.0. 1. Pyridoxal 5¢-phosphate (PLP). 2. 50 mM Sodium phosphate buffer pH 6.5. 3. Amicon® Ultra 0.5  mL centrifugal filter devices (5,000 MWCO) (Millipore). 4. The S-protein was purified from RNAse S (grade XII-S) (Sigma) using reversed-phase HPLC as described previously in ref. 9. 5. Protein G¢ (Invitrogen).

406

Lempens, Helms, and Merkx

2.10. Peptide/Protein Ligation to AminooxyFunctionalized SPR Chips

1. SPR chip with aminooxy groups (see Subheading 2.7). 2. Peptides or proteins with an N-terminal aldehyde or ketone (see Subheading 2.8 or 2.9). 3. 100 mM anilinium acetate, pH 4.5.

3. Methods Native chemical ligation (NCL) is a chemoselective reaction that occurs spontaneously between a peptide or protein containing a C-terminal thioester and a peptide with an N-terminal cysteine residue under aqueous conditions at neutral pH. The method was originally developed by Dawson et al. in 1994 to allow the chemical synthesis of proteins from smaller peptide fragments obtained by solid-phase peptide synthesis (10). The versatility of NCL was extended by the development of expression systems that use selfcleavable intein domains to generate recombinant proteins with C-terminal thioesters (e.g., IMPACT™ system from New England Biolabs (11)). In other application areas, NCL has been used for the conjugation of recombinant proteins to spectroscopic labels, dendrimers, liposomes, and micelles, as well as the site-specific immobilization of proteins to surfaces and the facile incorporation of unnatural amino acids (7, 12). In order to apply NCL for use in SPR, we developed a method that allows the functionalization of the SPR chip surface with cysteine residues with a free N-terminus (Fig. 2). Most commercially available SPR chips consist of a gold substrate that is coated with a nonfouling, carboxymethylated dextran layer. This dextran hydrogel surface layer can be readily modified to become reactive toward amine-containing compounds through the activation of the carboxylic acid groups by EDC/NHS treatment. To allow the functionalization of the carboxymethylated dextran layer with cysteine molecules using EDC/NHS coupling, we developed the thiazolidine derivative 4, which consists of a small diamine linker coupled to a nascent N-terminal cysteine. The thiazolidine ring protects the cysteine during the coupling reaction, but can be readily removed by treatment with 250  mM methoxyamine in buffer at pH 4. These relatively mild conditions are compatible with the Biacore’s microfluidics system components, thus allowing deprotection of the cysteine residues to be performed directly within the instrument. A variety of strategies have been developed to prepare synthetic peptides with C-terminal thioesters. Peptides with thioesters are still most efficiently synthesized using t-Boc chemistry (13), although recently, several new methods based on Fmoc chemistry (14) have also been reported. Recombinant full-length proteins

Chemoselective Protein and Peptide Immobilization on Biosensor Surfaces

407

Fig. 2. Synthetic strategies for the generation of cysteine- and aminooxy-functionalized SPR sensor chip surfaces starting from a carboxymethylated dextran-coated gold surface.

with a C-terminal thioester can be obtained using the commercially available IMPACT system from New England Biolabs, in which the gene of interest is cloned into an expression vector that generates a fusion protein with a C-terminal intein domain followed by a chitin domain. The chitin domain allows for the convenient purification of the fusion protein using chitin resins. Treatment with 50 mM MESNA induces the intein-catalyzed cleavage of the fusion protein, and then the protein of interest containing a C-terminal MESNA thioester is eluted from the chitin column. The expression of proteins whose stability depends on the pre­ sence of disulfide bonds can be cumbersome using the IMPACT system because of the use of reducing thiols during intein cleavage; however, two methods have recently been reported that allow efficient refolding using a MESNA-diMESNA redox couple (15) or allow efficient expression of the intein-fusion proteins by directing them to the bacterial periplasm (16). The second bioconjugation method described in this chapter is based on the formation of an oxime bond upon reaction of an aminooxy group with an aldehyde or ketone group. This strategy is complementary to NCL, as it allows chemoselective covalent immobilization of proteins via their N-terminus. The approach

408

Lempens, Helms, and Merkx

takes advantage of two recent advances in the field of oxime chemistry. Francis and coworkers have shown that treatment of proteins with pyridoxal 5¢-phosphate (PLP) allows site-specific oxidation of N-terminal amino acids into N-terminal ketones (17). Unlike the classical oxidation method using NaIO4, which is limited to N-terminal serine residues (18), PLP oxidation can be applied to most N-terminal amino acids, although with varying yields (19). Oxime ligations can thus be applied to a broad range of proteins without the need to introduce additional tags, provided that the N-terminus is not blocked by, for example, acylation. Recently, recombinant procedures have been developed by the Bertozzi group that allow site-specific introduction of formylglycine functionalities at any site within a protein sequence (20). The second important development is the finding by Dawson and coworkers that oxime ligations can be significantly accelerated using aniline as an organocatalyst (21). The latter finding has made oxime ligations one of the most efficient ligation reactions known today. Similar to the strategy used to generate cysteine-functionalized SPR chip surfaces, a protected aminooxy group linked to a small diaminoethyl spacer (8) can be used to functionalize commercially available carboxymethyl group-containing chips with aminooxy groups (Fig. 2). In the first step of the derivatization procedure, the free amine group of the t-Boc-protected aminooxy linker is conjugated to the carboxymethylated dextran surface using standard EDC/NHS chemistry. Deprotection of the aminooxy group in this case requires the removal of the chip from the Biacore instrument and overnight incubation of the chip in 1  M phosphoric acid buffer at pH 2 (22). When catalyzed by aniline, oxime ligations compare favorably to NCL with respect to reaction speed. A second practical advantage of oxime ligations is the absence of redox chemistry, making the method particularly attractive for the immobilization of disulfide bond-containing proteins such as immunoglobulins. 3.1. Synthesis of Thiazolidine-Protected Cysteine Linker (4)

The synthesis of the protected cysteine derivative involves a threestep process as described below (Fig. 3).

Fig. 3. Synthesis of a thiazolidine-protected cysteine.

Chemoselective Protein and Peptide Immobilization on Biosensor Surfaces 3.1.1. Preparation of N-(tert-Butyloxycarbonyl) Thiazolidine-4-Carboxylic Acid N-Hydroxysuccinimidyl Ester (1)

409

1. Dissolve N-(tert-butyloxycarbonyl)thiazolidine carboxylic acid (1.00 g, 4.3 mmol), NHS (542 mg, 4.7 mmol), HBTU (1.63  g, 4.3  mmol), and DIPEA (1.66  g, 12.9  mmol) in DMF (10 mL). 2. Stir the reaction mixture for 12 h. 3. Concentrate the reaction mixture to dryness in vacuo and dissolve the residue in ether (100 mL). 4. Wash the ethereal layer with saturated KCl (5 × 100  mL), saturated NaHCO3 (5 × 100  mL), and deionized water (10 × 100 mL). 5. Dry over Na2SO4, filter, and concentrate. The yield is 950 mg (67%). 1H NMR (CDCl3): (3:1 mixture of diastereomers): 5.16 (br, 1H, minor), 4.91 (t, J = 5.7  Hz, 1H, major), 4.61 (dd, J = 54, 9.2  Hz, 2H, major), 4.54 (dd, J = 65, 9.0 Hz, 2H, minor), 3.50 (m, 2H, minor), 3.41 (m, 2H, major), 2.83 (s, 4H, CH2CH2), 1.47 (s, 9H).

3.1.2. Preparation of N-(tertButyloxycarbonyl) Thiazolidine-4-Carboxylic Acid [2-(N-tertButyloxycarbonyl) Aminoethyl]Amide (3)

1. Add a solution of mono-N-(tert-butoxycarbonyl)ethylenediamine 2 (361 mg, 2.25 mmol) in DCM (2 mL) to a flask that contains 1 (676  mg, 2.05  mmol) and DIPEA (793  mg, 6.14 mmol) in DCM (3 mL). 2. Stir for 6 h. 3. Remove the solvent in vacuo and dissolve the residue in ethyl acetate (50 mL). 4. Extract the product with saturated NaHCO3 (5 × 50  mL), deionized water (5 × 50 mL), and brine (50 mL). 5. Dry the organic layer over Na2SO4, filter, and concentrate. The yield is 735 mg (95%). 1H NMR (CDCl3): 6.92 (br, 1H), 4.92 (br, 1H), 4.61 (m, 2H), 4.38 (br, 1H), 3.36 (m, 2H), 3.23 (m, 4H), 1.47 (s, 9H), 1.43 (s, 9H).

3.1.3. Preparation of Thiazolidine-4Carboxylic Acid (2-Aminoethyl)Amide (4)

1. Dissolve compound 3 (730 mg, 1.94 mmol) in a mixture of DCM and TFA (10 mL, 1:1 (v/v)). 2. Stir for 3 h. 3. Concentrate the mixture in vacuo to dryness and dissolve the residue in deionized water (10 mL). 4. Wash the aqueous layer with DCM (3 × 25  mL) and lyophilize. The yield is 330 mg (97%). 1H NMR (d6-DMSO): 8.64 (s, 1H), 7.83 (br, 3H), 6.0–5.0 (br, 2H), 4.28 (dd, J = 15.4, 6.6 Hz, 2H), 4.22 (dd, J = 15.8, 6.2 Hz, 1H), 3.34 (m, 2H), 3.25–3.12 (m, 2H), 2.88 (m, 2H).

410

Lempens, Helms, and Merkx

3.2. Functionalization of SPR Sensor Chips with N-Terminal Cysteines

1. Use HBS-EP (pH 7.4) as the running buffer on a Biacore T100 instrument and set the flow rate to 10 mL/min. 2. Insert a CM5 sensor chip and immobilize the thiazolidine derivative 4 in flow channels 2 and 4 of the Biacore instrument at 25°C according to the procedures described in steps 3–5 below. 3. Inject a solution containing 1.0 M EDC and 1.0 M NHS for 7 min. 4. Inject a solution containing 250 mM of 4 in 50 mM borate buffer (pH 8.5) for 7 min. 5. Inject a solution of ethanolamine HCl (1.0 M) at pH 8.5 for 7 min. 6. Use flow channels 1 and 3 as reference channels, and immobilize ethanolamine HCl according to the procedures described in steps 7–8 below. 7. Inject a solution containing 1.0 M EDC and 1.0 M NHS for 7 min. 8. Inject twice a solution of ethanolamine HCl (1.0 M) at pH 8.5 for 7 min each. 9. Raise the temperature of the chip to 37°C, and set the flow rate in channels 1, 2, 3, and 4 to 5 mL/min. 10. Inject twice a solution containing 250  mM methoxyamine HCl in 50 mM acetate buffer (pH 4.0) for 70 min over all channels. 11. Cool the chip to 25°C and set the flow rate in all channels to 10 mL/min. 12. Regenerate the surface of all channels with 50 mM NaOH in 0.5 M NaCl using 10 pulses of 30 s each (see Note 2).

3.3. Synthesis of Peptides with a C-Terminal b-Mercaptopropionic Acid-Leucine Thioester

1. Preparation of a TAMPAL resin to yield a C-terminal b-Mercaptopropionic Acid-Leucine (MPAL) thioester (13): Activate t-Boc-Leu-OH (1.1  mmol) by HBTU (1.0  mmol; 0.5  M in DMF) in the presence of DIPEA (3  mmol) for 3–4 min. Couple t-Boc-Leu-OH to 0.25 mmol of 4-methylbenzhydrylamine (MBHA) resin for 10  min. Remove unbound amino acid by rinsing with a DMF flow wash (2 × 20  s). Remove the t-Boc group by a TFA treatment (2 × 1  min). Perform a second DMF flow wash (2 × 20  s). Activate 1.1  mmol of S-tritylmercaptopropionic acid with 1.0 mmol of HBTU in the presence of 3 mmol DIPEA and couple it for 30 min to the Leu-MBHA resin. Perform a DMF flow wash (2 × 20 s). Remove the protecting trityl group by the addition of TFA containing 2.5% (v/v) triisopropylsilane and 2.5% (v/v) H2O. Perform a DMF flow wash (2 × 20 s).

Chemoselective Protein and Peptide Immobilization on Biosensor Surfaces

411

2. Grow the peptide by manual t-Boc-mediated solid-phase peptide synthesis (23): Activate each t-Boc-amino acid (1.1 mmol) by HBTU (1.0 mmol) in the presence of DIPEA (3 mmol) for 3–4 min. Couple all activated amino acids for 10 min, except for Arg, Ser, and Asn, which require a coupling time of 20 min. Perform a DMF flow wash (2 × 20 s). Remove the t-Boc group by a TFA treatment. Perform a DMF flow wash (2 × 20 s). 3. After synthesis, wash the peptide with DCM and 50% (v/v) MeOH in DCM to remove all DMF and dry in vacuo. 4. Cleave the peptide from the resin and remove the protecting groups from the side chains using HF treatment over 1 h at 0°C with 4% (v/v) p-cresol (see Note 1). 5. Precipitate the peptide in ice-cold diethyl ether. 6. Dissolve the pellet in acetonitrile and lyophilize. 7. Purify the peptide using reversed-phase HPLC. 3.4. Synthesis of GFP with a C-Terminal MESNA Thioester

1. Transform E. coli BL21(DE3) cells with the pGFPX plasmid. 2. Inoculate 2 mL of LB medium containing 100 mg/L ampicillin with a single colony of transformed bacteria. 3. Incubate overnight in a shaking incubator (225 rpm) at 37°C and transfer the 2-mL seed culture into 200  mL of LB medium containing 100 mg/L ampicillin. 4. Monitor the optical density of the cell culture at l = 600 nm using an absorbance spectrophotometer. At OD600 = 0.5, lower the incubation temperature to 15°C and add 0.3 mM IPTG to induce protein expression. 5. Collect the cells after overnight expression at 15°C (250 rpm) by low-speed centrifugation (~3,000 × g), and resuspend the cell pellet in BugBuster (Novagen) lysis buffer containing 1  mL/mL Benzonase Nuclease and incubate for 20  min at 20°C. 6. Centrifuge the cell lysate at 40,000 × g for 45 min to obtain a clear supernatant. 7. Load the supernatant onto a 10-mL chitin column equilibrated with column buffer. 8. Wash the column with 10 volumes of column buffer to remove nonbinding and nonspecifically bound proteins. 9. Flush the column with 3 volumes of cleavage buffer, and then close the column outlet and incubate the column overnight at 20°C. 10. Elute the GFP containing a MESNA thioester using 1 volume of cleavage buffer and store at −80°C.

412

Lempens, Helms, and Merkx

3.5. Peptide/Protein Ligation to CysteineFunctionalized SPR Chips

1. Incubate the peptide (0.5–5  mM) or protein (0.1  mM) containing a C-terminal thioester in HBS-EP buffer (pH 7.4) containing 50 mM 4-mercaptophenylacetic acid and 10 mM TCEP for 30 min at 25°C. 2. Inject the mixture from step 1 over flow channels 1 and 2 (or 3 and 4) of the Biacore instrument at a flow rate of 5 mL/min. The injection time may be varied in order to obtain surfaces with different immobilization levels. (As an example, the immobilization of a streptavidin-binding peptide (SLLAHPQGGGMPAL) (24) and a GFP are shown in Figs.  4 and 5, respectively.)

Fig. 4. (a) Ligation of a streptavidin-binding peptide to a cysteine-modified CM5 SPR sensor chip: (1) injection of the peptide with an MPAL-thioester in the presence of 50 mM MPAA and 10 mM TCEP; and (2) surface regeneration with 30-s washes with a buffer containing 50 mM NaOH and 500 mM NaCl. (b). Binding of streptavidin to the peptide-modified surface: (3) injection of 2 mM streptavidin; and (4) surface regeneration. Reproduced with permission from ref. 4 © 2007 Wiley-VCH Verlag GmbH & Co. KGaA.

Fig. 5. Immobilization of green fluorescent protein (GFP) to a cysteine-modified CM5 SPR sensor chip. The sample channel was functionalized with cysteines, whereas the reference channel was reacted with ethanolamine. 0.1 mM GFPMESNA thioester with 50 mM MPAA and 10 mM TCEP was injected over both channels for 1 h (1), followed by injection of buffer only. The difference in the signal response between the sample and reference channels represents the immobilization of 480 resonance units (RU) of GFP. The inset shows the specific binding of anti-GFP antibody JL-8 (Clontech) to the sensor surface of the sample channel (2). Reproduced with permission from ref. 4 © 2007 Wiley-VCH Verlag GmbH & Co. KGaA.

Chemoselective Protein and Peptide Immobilization on Biosensor Surfaces

413

3. Regenerate the surfaces of all the channels using 30-s pulses of an appropriate buffer solution until a stable baseline is reached. 4. Perform binding experiments on the peptide- or proteinmodified sensor surface (see Figs. 4 and 5, for examples; also see Note 3). The synthesis of the protected aminooxy derivative involves a three-step process as described below (Fig. 6).

3.6. Synthesis of t-Boc-Protected Aminooxy Linker (8)

1. Add benzyl phenyl carbonate (1.00 g, 4.4 mmol) to a stirred solution of ethylenediamine (263  mg, 4.4  mmol) in EtOH (17 mL).

3.6.1. Preparation of (2-Aminoethyl)Carbamic Acid Benzyl Ester (6)

2. Stir the reaction mixture for 12 h. 3. Concentrate the reaction mixture in vacuo and dissolve the residue in H2O (25 mL). 4. Adjust the pH to 3 by the addition of aqueous HCl (3 M) and extract the aqueous layer with DCM (2 × 50 mL). 5. Adjust the pH of the aqueous layer to 11 by the addition of aqueous NaOH (5 M) and extract with DCM (3 × 80 mL). 6. Dry the organic layer over Na2SO4, filter, and concentrate. The yield is 492  mg (58%). 1H NMR (CDCl3): 7.37–7.33 (m, 5H), 5.11 (br s, 3H), 3.26–3.24 (m, 2H), 2.83 (t, 2H, J = 5.9 Hz), 1.38 (br s, 2H). 1. Dissolve 6 (272  mg, 2.7  mmol), N-[(tert-butyloxycarbonyl) aminooxy]acetic acid (268 mg, 2.7 mmol), HCTU (609 mg, 2.8 mmol), and DIPEA (572 mg, 8.5 mmol) in DMF (5 mL).

3.6.2. Preparation of 2-[(tertButyloxycarbonyl) Aminooxy]-N[2(Benzyloxycarbonyl Amino) Ethyl]Acetamide ( 7)

2. Stir the reaction mixture for 12 h. 3. Remove the solvent in vacuo and dissolve the residue in ethyl acetate (50 mL).

O H2N

+

NH2

H N

O

O

O

NH2

O 6

5

O

O N H

O

O

N H

O 8

O

O

O

O

N H

NH2

Fig. 6. Synthesis of a protected aminooxy linker.

O

N H

O 7

N H

H N

O

O O

OH

414

Lempens, Helms, and Merkx

4. Extract with saturated KCl (3 × 50 mL), saturated NaHCO3 (3 × 50 mL), deionized water (50 mL), and brine (50 mL). 5. Dry the organic layer over Na2SO4, filter, and concentrate. The yield is 418 mg (81%). 1H NMR (CDCl3): 8.21 (br s, 1H), 7.74 (s, 1H), 7.37–7.32 (m, 5H), 5.45 (br s, 1H), 5.11 (s, 2H), 4.34 (s, 2H), 3.45–3.41 (m, 2H), 3.41–3.38 (m, 2H), 1.42 (s, 9H). 3.6.3. Preparation of 2-[(tertButyloxycarbonyl) Aminooxy]-N(2Aminoethyl)Acetamide (8)

1. Add Pd/C catalyst (30  mg) to a solution of 7 (300  mg, 0.82 mmol) in ethanol (25 mL). 2. Degas the suspension and set it under a H2-atmosphere (1 bar) for 4 h. 3. After exposure to H2, filter the reaction mixture through Celite® and concentrate in vacuo. 4. Dissolve the residue in deionized water (30 mL), extract with DCM (3 × 30 mL), and lyophilize. The yield is 170 mg (89%). 1H NMR (d6-DMSO): 4.17 (s, 2H), 3.35–3.30 (m, 2H), 2.84 (t, 2H, J = 6.2 Hz), 1.42 (s, 9H).

3.7. Functionalization of SPR Sensor Chips with Aminooxy Groups

1. Use HBS-EP (pH 7.4) as the running buffer on a Biacore T100 instrument and set the flow rate to 10 mL/min. 2. Insert a CM5 chip and immobilize the aminooxy derivative 8 in flow channels 2 and 4 at 25°C according to the procedures described in steps 3–5 below. 3. Inject a solution containing 1.0 M EDC and 1.0 M NHS for 7 min. 4. Inject a solution containing 250 mM of 8 in 50 mM borate buffer (pH 8.5) for 7 min. 5. Inject a solution of ethanolamine HCl (1.0 M) at pH 8.5 for 7 min. 6. Use flow channels 1 and 3 as reference channels and immobilize ethanolamine HCl according to the procedures described in steps 7–8 below. 7. Inject a solution containing 1.0 M EDC and 1.0 M NHS for 7 min. 8. Inject twice a solution of ethanolamine HCl (1.0 M) at pH 8.5 for 7 min each. 9. Eject the sensor chip from the Biacore instrument docking area. 10. Incubate the sensor chip overnight in phosphoric acid buffer (1 M) at pH 2.0. 11. Rinse the sensor chip surface with deionized water and re-insert the chip into the Biacore instrument. 12. Regenerate the surface of all channels using 2 × 30-s pulses of HCl (100 mM), NaOH (50 mM), and 0.5% (w/v) SDS at a flow rate of 100 mL/min.

Chemoselective Protein and Peptide Immobilization on Biosensor Surfaces

3.8. Synthesis of Peptides or Proteins with an N-Terminal Aldehyde Group

415

1. Synthesize the peptide using either Fmoc- or t-Boc-mediated solid-phase peptide synthesis. 2. Couple an extra serine residue at the N-terminus of the peptide before cleavage. 3. Treat the purified peptide for 5 min with 1.2 equivalents of NaIO4 in 0.01 M sodium phosphate buffer, pH 7.0, at 4°C (18). For proteins that contain a serine or threonine at their N-terminus, this step can be performed to introduce an aldehyde as well. 4. Purify the peptide after 5 min by reversed-phase HPLC.

3.10. Peptide/Protein Ligation to AminooxyFunctionalized SPR Chips

1. Dissolve the protein (33 mM) together with PLP (6.7 mM) in sodium phosphate buffer (50 mM, pH 6.5) (17). 2. Incubate overnight at 41°C. 3. Remove the excess PLP remaining via repeated concentration and dilution in 50  mM sodium phosphate buffer (pH 6.5) using Amicon® centrifugal filter devices. 1. Dissolve the peptide (1 mM) or protein (5–100 mM) containing an N-terminal aldehyde or ketone group in 100  mM anilinium acetate, pH 4.5. If the protein is not stable at pH 4.5, then buffers with higher pH values can be used. In the presence of p-methoxyaniline as a catalyst, efficient immobilization can still be reached at these pH values (Fig. 7). 500 S-peptide Immobilization / RU

3.9. Synthesis of Peptides or Proteins with an N-Terminal Ketone Group (see Notes 4 and 5)

pH 4.5 catalyzed pH 6.0 catalyzed pH 7.0 catalyzed pH 4.5 uncatalyzed pH 6.0 uncatalyzed pH 7.0 uncatalyzed

400 300 200 100 0

1

2

Cycle

3

4

Fig.  7. Efficiency of S-peptide immobilization to SPR sensor chips under various pH conditions and in the presence or absence of catalytic aniline derivatives. Referencesubtracted immobilization levels are shown for an aminooxy SPR sensor chip surface functionalized using sequential injections (4 × 12 s) of 1 mM NaIO4-treated S-peptide in various buffer solutions (100 mM anilinium acetate, pH 4.5; 100 mM anisidine in HBS-EP, pH 6.0; 100  mM anisidine in HBS-EP, pH 7.0; 100  mM ammonium acetate, pH 4.5; HBS-EP, pH 6.0; or HBS-EP, pH 7.0) at room temperature. Reproduced with permission from ref. 5 © 2007 Wiley-VCH Verlag GmbH & Co. KGaA.

416

Lempens, Helms, and Merkx

i

iii

ii

iv

Response / RU

41000 40800 40600

130 RU

40400 40200 40000 6200

6400

6600

6800

7000

7200

Time / s Fig. 8. Surface immobilization of S-peptide to SPR sensor chips using catalyzed oxime ligation. A 12-s pulse of 1 mM NaIO4-treated S-peptide in 100 mM anilinium acetate at pH 4.5 (i) results in 130 RU of immobilized peptide after two regeneration steps using 10 mM glycine, pH 1.5 (ii), subsequent injection of 150 nM S-protein analyte (iii) results in specific binding to the S-peptide-functionalized channel surface. Finally, two regeneration steps using 10 mM glycine (pH 1.5) (iv) were performed to remove the noncovalently bound S-protein analyte from the surface. The sample channel surface (upper trace) was functionalized with aminooxy groups, whereas the reference channel surface (lower trace) was modified with ethanolamine. Reproduced with permission from ref. 5 © 2007 Wiley-VCH Verlag GmbH & Co. KGaA.

2. Inject the mixture from step 1 over flow channels 1 and 2 (or 3 and 4) at a flow rate of 10 mL/min. Vary the injection time to obtain surfaces with different immobilization levels (as examples, the immobilization of the S-peptide (aldehydeKETAAAKFERQHMDS) (25) and Protein G¢ are shown in Figs. 8 and 9, respectively). 3. Regenerate the surfaces of all the channels using 30-s pulses of an appropriate buffer solution until a stable baseline is reached (see Note 6). 4. Perform binding experiments on the peptide- or proteinmodified sensor surface (see Note 7).

4. Notes 1. Peptides with a C-terminal thioester prepared via t-Boc chemistry require the use of HF for cleavage. Caution: HF is an extremely corrosive and poisonous acid, and should be handled with extreme care. Alternatively, these peptides can also be prepared via Fmoc-mediated solid-phase peptide synthesis using, for example, safety-catch resins (14).

Chemoselective Protein and Peptide Immobilization on Biosensor Surfaces

417

Fig. 9. Comparison of CM5 SPR sensor chips functionalized with 20 RU of protein G¢ using oxime ligation (upper trace) versus classical amine coupling (lower trace). The chip surface obtained using classical amine coupling binds only 17 RU of IgG analyte under saturating concentrations, whereas 130 RU are expected assuming 1:1 binding. The same amount of protein G¢ immobilized via oxime ligation showed 97 RU of IgG analyte binding under identical conditions. Immobilization of 20 RU of protein G¢ required incubation with 5 mM PLP-treated protein G¢ (27% of the N-terminal methionines were oxidized to a ketone) in 100 mM anilinium acetate (pH 4.5) for only 36 s, whereas a 5-min injection of 1 mM protein G¢ was required to reach the same immobilization level using classical amine coupling. Reproduced with permission from ref. 5 © 2007 Wiley-VCH Verlag GmbH & Co. KGaA.

2. Sensor chips that were prepared according to the above procedure were used directly in ligation experiments or stored at 4°C in HBS-EP buffer (pH 7.4) until needed. For the chips prepared via native chemical ligation, TCEP (2  mM) was added to the buffer. 3. During binding experiments, we typically use buffers that include 2 mM of TCEP or DTT to prevent oxidation of the surface cysteines. In the absence of these reducing agents, we occasionally observed some baseline drift, which we ascribed to the formation of disulfide bonds. However, these reducing agents may sometimes interfere with proteins whose stability depends on the presence of disulfide bonds. To ensure a stable baseline in the absence of reducing agents, the free sulfhydryl groups can be reacted with DTNB (Ellman’s reagent) to form stable disulfide bonds. If required, DTNB protection can be removed by treatment with DTT or TCEP. 4. Unlike the oxidation of N-terminal serines by NaIO4, the oxidation of N-terminal amino acids by PLP is not quantitative. Yields vary between 10 and 80%, and depend strongly on the specific amino acids present. For SPR applications, these nonstoichiometric yields typically do not present a serious

418

Lempens, Helms, and Merkx

problem, since nonoxidized proteins will not react to the sensor chip surface and will be removed by the flow system. Moreover, proteins containing Q, W, H, P, and K at the N-terminus have been reported to be essentially unreactive toward PLP oxidation (19). 5. The conversion of N-terminal amino acids into N-terminal ketones during PLP oxidation can be quantified by the siteselective attachment of PEG polymers followed by SDSPAGE analysis (17). 6. Unlike NCL, oxime bond formation is in principle reversible. However, no significant decrease in immobilization levels were observed even after overnight incubation of the chip under a constant flow of buffer, which is consistent with the fact that the oxime bonds are very stable under neutral conditions (26). Partial cleavage of the oxime bond is possible, but only under extreme, nonphysiological conditions such as a constant flow of buffer containing 250 mM methoxylamine and 100 mM anilinium acetate at pH 4.5. 7. Immobilization of protein G¢ can also be performed using C1 chips that lack the dextran layer and are thus more similar to many other non-SPR chip surfaces (5).

Acknowledgments The authors would like to thank Dr. Ingrid van Baal and Professor E.W. Meijer for co-developing the methods described in this chapter. This work was supported by a VIDI grant from the Netherlands Organization of Scientific research (NWO; VIDI 700 56.428). References 1. a) Johnsson, B., Lofas, S., Lindquist, G. (1991) Immobilization of proteins to a ­carboxymethyldextran-modified gold surface for biospecific interaction analysis in surface plasmon resonance sensors. Anal. Biochem. 198, 268–277; b) Peluso, P., Wilson, D. S., Do, D., Tran, H., Venkatasubbaiah, M., Quincy, D., Heidecker, B., Poindexter, K., Tolani, N., Phelan, M., Witte, K., Jung, L. S., Wagner, P., Nock, P. (2003) Optimizing antibody immobilization strategies for the construction of protein microarrays. Anal. Biochem. 312, 113–124; c) Kortt, A. A., Oddie, G. W., Iliades, P., Gruen, L. C., Hudson, P. J. (1997) Nonspecific amine immobilization of ligand can be a potential

source of error in BIAcore binding experiments and may reduce binding affinities. Anal. Biochem. 253, 103–111. 2. Jonkheijm, P., Weinrich, D., Schroeder, H., Niemeyer, C. M., Waldmann, H. (2008) Chemical strategies for generating protein biochips. Angew. Chem. Int. Ed. 47, 9618–9647. 3. a) Girish, A., Sun, H., Yeo, D. S. Y., Chen, G. Y. J., Chua, T.-K., Yao, S. Q. (2005) Sitespecific immobilization of proteins in a microarray using intein-mediated protein splicing. Bioorg. Med. Chem. Lett. 15, 2447–2451; b) Houseman, B. T., Huh, J. H., Kron, S. J., Mrksich, M. (2002) Peptide chips for the quantitative evaluation of protein kinase

Chemoselective Protein and Peptide Immobilization on Biosensor Surfaces ­activity. Nat. Biotechnol. 20, 270–274; c) Kalia, J., Abbott, N. L., Raines, R. T. (2007) General method for site-specific protein immobilization by staudinger ligation. Bioconjug. Chem. 18, 1064–1069; d) Camarero, J. A., Kwon, Y., Coleman, M. A. (2004) Chemoselective attachment of biologically active proteins to surfaces by expressed protein ligation and its application for “protein chip” fabrication. J. Am. Chem. Soc. 126, 14730–14731. 4. Helms, B., van Baal, I., Merkx, M., Meijer, E. W. (2007) Site-specific protein and peptide immobilization on a biosensor surface by pulsed native chemical ligation. ChemBioChem 8, 1790–1794. 5. Lempens, E. H. M., Helms, B. A., Merkx, M., Meijer, E. W. (2009) Efficient and chemoselective surface immobilization of proteins by using aniline-catalyzed oxime chemistry. ChemBioChem 10, 658–662. 6. Jones, D. S. Coutts, S. M., Gamino, C. A., Iverson, G. M., Linnik, M. D. Randow, M. E., Ton-Nu, H.-T., Victoria, E. J. (1999) Multivalent thioether-peptide conjugates: B cell tolerance of an anti-peptide immune response. Bioconjug. Chem. 10, 480–488. 7. van Baal, I., Malda, H., Synowsky, S. A., van Dongen, J. L. J., Hackeng, T. M., Merkx, M., Meijer, E. W. (2005) Multivalent peptide and protein dendrimers using native chemical ligation. Angew. Chem. Int. Ed. 44, 5052–5057. 8. Pittelkow, M., Lewinsky, R., Christensen, J. B. (2002) Selective synthesis of carbamate protected polyamines using alkyl phenyl carbonates. Synthesis 15, 2195–2202. 9. Lempens, E. H. M., van Baal, I., van Dongen, J. L. J., Hackeng, T. M., Merkx, M., Meijer, E. W. (2009) Noncovalent synthesis of protein dendrimers. Chem. Eur. J. 15, 8760–8767. 10. a) Dawson, P. E., Muir, T. W., Clark-Lewis, I., Kent, S. B. H. (1994) Synthesis of proteins by native chemical ligation. Science 266, 776– 779; b) Johnson, E. C. B., Kent, S. B. H. (2006) Insights into the mechanism and catalysis of the native chemical ligation reaction. J. Am. Chem. Soc. 128, 6640–6646. 11. Muir, T. W., Sondhi, D., Cole, P. A. (1998) Expressed protein ligation: a general method for protein engineering. Proc. Natl. Acad. Sci. USA 95, 6705–6710. 12. a) Reulen, S. W. A., Brusselaars, W. W. T., Langereis, S., Mulder, W. J. M., Breurken, M., Merkx, M. (2007) Protein-liposome conjugates using cysteine-lipids and native chemical ligation. Bioconjug. Chem. 18, 590–596; b) Reulen, S. W. A., Dankers, P. Y. W., Bomans, P. H. H., Meijer, E. W., Merkx, M. (2009)

419

Collagen targeting using protein-­functionalized micelles: the strength of multiple weak interactions. J. Am. Chem. Soc. 131, 7304–7312. 13. Hackeng, T. M., Griffin, J. H., Dawson, P. E. (1999) Protein synthesis by native chemical ligation: expanded scope by using straightforward methodology. Proc. Natl. Acad. Sci. USA 96 10068–10073. 14. a) Ingenito, R., Bianchi, E., Fattori, D., Pessi, A. (1999) Solid phase synthesis of peptide C-terminal thioesters by Fmoc/t-Bu chemistry. J. Am. Chem. Soc. 121, 11369–11374; b) Blanco-Canosa, J. B., Dawson, P. E. (2008) An efficient Fmoc-SPPS approach for the generation of thioester peptide precursors for use in native chemical ligation. Angew. Chem. Int. Ed. 47, 6851–6855. 15. Bastings, M. M. C., van Baal, I., Meijer, E. W., Merkx, M. (2008) One-step refolding and purification of disulfide-containing proteins with a C-terminal MESNA thioester. BMC Biotechnology 8, 76. 16. Reulen, S. W. A., van Baal, I., Raats, J. M. H., Merkx, M. (2009) Efficient, chemoselective synthesis of immunomicelles using singledomain antibodies with a C-terminal thioester. BMC Biotechnology 9, 66. 17. Gilmore, J. M., Scheck, R. A., Esser-Kahn, A. P., Joshi, N. S., Francis, M. B. (2006) N-terminal protein modification through a biomimetic transamination reaction. Angew. Chem. Int. Ed. 45, 5307–5311. 18. Geoghegan, K. F., Stroh, J. G. (1992) Sitedirected conjugation of nonpeptide groups to peptides and proteins via periodate oxidation of a 2-amino alcohol. Application to modification at N-terminal serine. Bioconjug. Chem. 3, 138–146. 19. Scheck, R. A., Dedeo, M. T., Iavarone, A. T., Francis, M. B. (2008) Optimization of a biomimetic transamination reaction. J. Am. Chem. Soc. 130, 11762–11770. 20. a) Carrico, I. S., Carlson, B. L., Bertozzi, C. R. (2007) Introducing genetically encoded aldehydes into proteins. Nat. Chem. Biol. 3, 321– 322; b) Rush, J. S., Bertozzi, C. R. (2008) New aldehyde tag sequences identified by screening formylglycine generating enzymes in vitro and in vivo. J. Am. Chem. Soc. 130, 12240–12241. 21. Dirksen, A., Hackeng, T. M., Dawson, P. E. (2006) Nucleophilic catalysis of oxime ligation. Angew. Chem. Int. Ed. 45, 7581–7584. 22. Li, B., Bemish, R., Buzon, R. A., Chiu, C. K.-F., Colgan, S. T., Kissel, W., Le, T., Leeman, K. R., Newell, L., Roth, J. (2003) Aqueous phosphoric acid as a mild reagent for

420

Lempens, Helms, and Merkx

­ eprotection of the tert-butoxycarbonyl group. d Tetrahedron Lett. 44, 8113–8115. 23. Schnolzer, M., Alewood, P., Jones, A., Alewood, D., Kent, S. B. (1992) In situ neutralization in Boc-chemistry solid phase peptide synthesis. Rapid, high yield assembly of difficult sequences. Int. J. Peptide Protein Res. 40, 180–93. 24. Bastings, M. M. C., Helms, B. A., van Baal, I., Hackeng, T. M., Merkx, M, Meijer, E. W. (2011) From phage display to dendrimer

display: insights into multivalent binding 133, doi:10.1021/ja110700x. 25. a) Richards, F. M., Vithayathil, P. J. (1959) Preparation of subtilisin-modified ribonuclease and the separation of the peptide and protein components. J. Biol. Chem. 234, 1459-1465; b) Raines, R. T. (1998) Ribonuclease A. Chem. Rev. 98, 1045–1065. 26. Kalia, J., Raines, R. T. (2008) Hydrolytic stability of hydrazones and oximes. Angew. Chem. Int. Ed. 47, 7523–7526.

Chapter 26 Fabrication of Dynamic Self-Assembled Monolayers for Cell Migration and Adhesion Studies Nathan P. Westcott and Muhammad N. Yousaf Abstract How cells interact with the extracellular matrix (ECM) is important for a number of fundamental ­processes in cell biology. However, the ECM is highly complex and in order to simplify the matrix for cell biological studies, it has been modeled with self-assembled monolayers (SAMs) of alkanethiolates on gold substrates. In this chapter, we outline procedures to create dynamic surfaces by functionalizing SAMs. SAMs based on quinone, oxyamine, and alcohol-terminated thiols were used to immobilize cell adhesive peptides with spatial control. Cells were seeded to these surfaces to provide cell co-culture ­patterns suitable for biological studies. Key words: Self-assembled monolayers, Alkanethiolates, Soft lithography, Photochemistry, Cell migration, Dynamic surface, Cell adhesion

1. Introduction The interaction of cells and the ECM is important for a number of fundamental processes, including cancer metastasis, cell migration, and cell–cell communication (1). However, the ECM is a complex mixture of proteins, peptides, and small molecules, making the cell–matrix interaction difficult to study. In order to model the ECM, several different materials have been developed, including polymers and SAMs (2). In particular, SAMs of alkanethiolates on gold have been extensively studied (3). SAMs have several inherent advantages when serving as a model for the ECM. They are synthetically flexible, compatible with a large number of surface spectroscopic techniques, and have the ability to be made inert toward nonspecific adsorption and adhesion. One of the most common applications of SAMs in biology is cell patterning. Currently, microcontact printing has been the Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_26, © Springer Science+Business Media, LLC 2011

421

422

Westcott and Yousaf

method of choice to pattern cells (4). For example, the technique has been used to control cell shape, which has been shown to have an effect on the cell’s subsequent migration (5). However, this technique is limited to printing hydrophobic patterns on SAMs, followed by backfilling with an inert background. Typically, proteins are adsorbed to hydrophobic patches and cells interact with the resulting surface. Other techniques have been developed using electrochemical desorption (6) and gold etching (7) to pattern cells on hydrophobic regions. However, these techniques require complex instruments, whereas – ideally – cell patterning procedures should be simple so that the broadest audience possible can utilize them. To that end, we have developed a number of different bioimmobilization strategies based on SAMs to create dynamic substrates for cell patterning. Utilizing a combination of microfluidics, photolithography, and SAMs, biomolecules were patterned to study cell adhesion and migration without interference from nonspecific protein adsorption. In this chapter, we outline procedures to functionalize SAMs for biological applications based on SAMs terminated with alcohol (8), quinone (9), and oxyamine functionalities (10). These surfaces were then patterned with a cell adhesive peptide to generate cell co-culture patterns.

2. Materials 2.1. General Procedures

1. Falcon™ evaporation-proof Petri dishes (50 × 9  mm, TightFit Lid) (BD Biosciences, San Jose, CA).

2.1.1. General Reagents

2. Dimethylformamide (DMF). 3. Ethanol. 4. Acetonitrile. 5. Dichloromethane (DCM). 6. Perchloric acid. 7. Sulfuric acid. 8. Trifluoroacetic acid. 9. Phosphate-buffered saline (PBS). 10. Petri dishes (35 × 10 and 60 × 15 mm). 11. Polyethylene tubes (5 and 50 mL).

2.1.2. Photolithography

1. Illustrator CS3 software (Adobe, San Jose, CA). 2. Custom-designed photomask (PageWorks, Cambridge, MA). 3. Sylgard 184 Polydimethylsiloxane (PDMS) kit (Dow Corning, Greensborough, NC). 4. Silicon wafers (4-in. diameter, á100ñ) (Silicon Sense, Nashua, NH).

Fabrication of Dynamic Self-Assembled Monolayers

423

5. Quartz plate (4 × 4 in.). 6. SU-8 150 photoresist (Microchem, Newton, MA). 7. Blak-Ray® High Intensity UV lamp (365  nm wavelength) (UVP, Upland, CA). 8. SU-8 developer (Microchem). 9. Soxhlet extractor (Chemglass). 10. 53/4-in. Glass pipettes. 11. KW-4A Spin Coater (Chemat, Northridge, CA). 12. Perfluoroheptadecyl trichlorosilane Technologies, Bristol, PA).

(United

Chemical

13. X-Acto knife. 2.1.3. Gold Thin Film Deposition

1. Fisherfinest™ glass microscope slides (1 × 3 in.) (Fisher). 2. Hydrogen peroxide, 30% (v/v). 3. Electron-beam gold evaporator tool (Thermionics, Port Townsend, WA). 4. 99.99% Gold Slug (Sigma, St. Louis, MO). 5. 99.99% Titanium (Sigma).

2.1.4. SAM Formation

1. Gold-evaporated glass slides. 2. Glass cutter. 3. 1 mM thiol solution in ethanol.

2.1.5. (6-Nitroveratryl) oxycarbonyl Photodeprotection

1. High-intensity UV lamp (365 nm). 2. (6-Nitroveratryl)oxycarbonyl (NVOC)-protected SAM. 3. Photomask (Pageworks, Cambridge, MA). 4. Quartz plate (4 × 4 in.).

2.1.6. Synthesis of Ketone- and OxyamineTerminated Peptides

1. Fluorenylmethyloxycarbonyl (Fmoc)-protected arginine, ­serine, glycine, and aspartate (Anaspec, Fremont, CA). 2. Triisopropylsilane. 3. Rink amide resin (Anaspec). 4. Fmoc-aminocaproic acid (Anaspec). 5. Butoxycarbonyl (Boc) 3-(aminooxy)acetic acid (Anaspec). 6. CS136XT Automated Peptide Synthesizer (CS Bio Co, Menlo Park, CA). 7. 4-Acetylbutyric acid. 8. FreeZone 2.5 Lyophilizer (Labconco, Kanas City, MS). 9. Peptide Synthesis Vessel (Chemglass).

424

Westcott and Yousaf

2.1.7. Cell Culture

1. Dulbecco’s Modified Eagle’s Medium (DMEM) (Gibco, Bethesda, MD). 2. Serum containing medium: 10% (v/v) calf bovine serum (MP Biomedicals, Solon, OH), 1% (v/v) 200× streptomycin/penicillin solution (MP Biomedicals), and 94% (v/v) DMEM. 3. Cell culture flasks (100 cm2) (Dow Corning). 4. 3T3-Swiss albino fibroblast cells. 5. Trypsinization solution: 0.05% (v/v) Trypsin and 0.053 mM EDTA in DMEM (MP Biomedicals). 6. Cell incubator (Ultima II, Fisher).

2.1.8. Cell Seeding

1. SAM surface. 2. DMEM (Gibco). 3. Serum-containing medium.

2.1.9. Peptide Immobilization

1. Parafilm. 2. 10 mM peptide solution (2 or 3, Fig. 1). 3. Activated SAM surface.

Fig. 1. Chemical structures of the molecules used in these procedures.

Fabrication of Dynamic Self-Assembled Monolayers

2.2. Quinone Cell Patterning Strategy 2.2.1. Chemical Surface Activation

2.2.2. Electrochemical Surface Activation

425

1. Saturated benzoquinone in ethanol. 2. 1 mM tetra(ethylene glycol) undecane thiol (TEG) (1, Fig. 1) in ethanol. 3. 1  mM photoprotected hydroquinone-terminated thiol (4, Fig. 1) in ethanol. 1. 1 M Perchloric acid. 2. Epsilon potentiostat (Bioanalytical Systems, West Lafayette, IN). 3. Ag/AgCl reference electrode (Bioanalytical Systems). 4. Coiled platinum wire counter electrode (Bioanalytical Systems). 5. Photodeprotected hydroquinone SAM substrate.

2.2.3. Electrochemical Ligand Readout

1. Oxime-functionalized SAM substrate.

2.2.4. Cell Co-culture

1. 3T3-Swiss albino fibroblast cells.

2. Electrochemical potentiostat and setup from Subheading 2.2.2.

2. Transfected 3T3-Swiss albino fibroblast cells expressing green fluorescent protein (GFP)-actin. 3. 10 mM GRGDS–oxyamine in serum-containing media. 2.3. Oxyamine Cell Patterning Strategy

1. 10 mM Ketone-terminated GRGDS peptide. 2. Photopatterned oxyamine surface. 3. Saturated semicarbazide solution. 4. 1  mM tetra(ethylene glycol) undecane thiol (1, Fig.  1) in ethanol. 5. 1 mM photoprotected oxyamine-terminated thiol (6, Fig. 1) in ethanol.

2.4. Aldehyde Cell Patterning Strategy

1. 60  mM Pyridinium chlorochromate (PCC, 8, Fig.  1) in acetonitrile. 2. 10 mM oxyamine-terminated GRGDS peptide. 3. 1  mM tetra(ethylene glycol) undecane thiol (1, Fig.  1) in ethanol.

3. Methods 3.1. General Procedures 3.1.1. Use of Reagents

1. All reagents were used as received unless otherwise noted. 2. Caution: Take care when handling concentrated acids. Wear chemical protective clothing (gloves and aprons) and safety goggles.

426

Westcott and Yousaf

3.1.2. Photolithography

Photolithography is used to generate PDMS stamps and microfluidic cassettes for patterning SAM surfaces. 1. Design the photomask patterns for creating the PDMS stamps and microfluidic cassettes in Illustrator software (see Notes 1 and 2), and submit the designs to a photomask printing facility (e.g., PageWorks). 2. To fabricate the PDMS stamps and microfluidic cassettes using photolithography, first start with a 4-in. silicon wafer and pour 4 mL of SU-8 150 onto the wafer. 3. Place the wafer on a spin coater and spin at 500 RPMs for 30 s, then at 1,500 RPMs for 90 s. 4. Prebake the photoresist on a hot plate at 65°C for 10 min, and then at 95°C for 1 h (see Note 3). 5. After the photoresist has been baked, create a sandwich with the wafer on the bottom, the photomask in the middle, and a quartz plate on top. 6. Place the sandwich under a UV lamp and expose for 60 s. 7. Remove the quartz plate and photomask from the silicon wafer. 8. Postbake the photoresist on a hot plate at 65°C for 1 min, and then at 95°C for 20 min (see Note 4). 9. After postbaking, place the wafer in SU-8 developer solution and stir for 10  min or until the noncrosslinked SU-8 is removed. 10. Rinse the SU-8-patterned wafer in ethanol and dry under vacuum. 11. After the wafer and photoresist are dry, place the wafer inside a vacuum chamber with a bottle of perfluoroheptadecyl trichlorosilane. Open the cap of the perfluoroheptadecyl trichlorosilane bottle and apply a vacuum to the chamber for 1  h. Afterward, remove the silicon wafer (master mold) from the vacuum chamber. 12. Prepare a mixture containing PDMS base:curing agent in a 10:1 (w/w) ratio. Pour the PDMS prepolymer mixture into a large Petri dish and stir vigorously. 13. Degas the PDMS prepolymer mixture under vacuum for 10 min, and then pour the degassed solution over the silicon master mold (from step 11). Cure the PDMS at 75°C for 1 h. 14. Remove the solid PDMS layer from the silicon master and slice out the PDMS stamps/microfluidic cassettes with an X-Acto knife. 15. For the microfluidic cassettes, punch out access holes with a glass pipette to allow for fluid flow.

Fabrication of Dynamic Self-Assembled Monolayers 3.1.3. Gold Thin Film Deposition

427

Gold is evaporated on glass slides to provide a substrate for SAM formation (see Note 5). The gold layer is thin to ensure transparency for microscopy. 1. Clean 8 or 16 glass slides by placing them in piranha solution (concentrated sulfuric acid: 30% (w/w), hydrogen peroxide, 3:1 v/v) for 4 h. Caution: Piranha solution reacts violently with most organic materials and must be handled with extreme care. Wear safety glasses and nitrile gloves and always work inside a fume hood. 2. Remove the glass slides from the piranha solution, rinse with water and then with ethanol, and dry under a stream of nitrogen. 3. Load the slides into the gold evaporator and turn on the vacuum pump. 4. Allow the pressure to reach <10−5  torr and then evaporate 6 nm of titanium and 15 nm of gold on the glass slides. The gold and titanium layers are thin so that they remain translucent for microscopy. 5. Allow the evaporator to reach atmospheric pressure and remove the gold slides from the holder.

3.1.4. SAM Formation

SAMs are formed by immersing a gold substrate in an ethanolic solution of thiol molecules. The resulting surface can then be further modified for cell patterning. 1. Cut a gold-coated glass slide into 1 × 2  cm2 pieces using a glass cutter. 2. Rinse the cut glass pieces with ethanol and place into an evaporation-proof Petri dish. 3. Add 5 mL of 1 mM thiol solution in ethanol and let the formation of SAMs on the gold-coated glass pieces to proceed for 12 h. (Fig. 2)

Fig. 2.  Schematic representation of a self-assembled monolayer of alkanethiolates on gold. When the gold substrates are immersed in an alkanethiol solution, the thiol groups bind to the gold while the alkane chains pack together, presenting their terminal headgroups at the surface.

428

Westcott and Yousaf

3.1.5. NVOC Photodeprotection

Photodeprotection unmasks the functional groups of NVOCprotected SAMs so that they can be reacted with peptides. 1. Make a sandwich with a photoprotected SAM substrate on the bottom, the photomask in the middle, and a quartz plate on top. 2. Illuminate the sandwich for 30  min under a high-intensity UV lamp (365 nm wavelength) by orienting the light output cable directly above and perpendicular to the SAM surface. 3. Remove the SAM surface from UV exposure and rinse with ethanol.

3.1.6. Peptide Synthesis

The RGD peptide is an epitope of fibronectin, an ECM protein. The different terminal groups allow for the peptide immobilization to different surface chemistries. 1. Weigh out 0.3 mol of Fmoc-protected serine, aspartic acid, arginine, resin, and aminocaproic acid. For Foc-protected glycine, weigh out 0.6 mol. 2. Weigh out 0.172 g of Rink amide resin. 3. For synthesis of the oxyamine peptide, weigh out 0.3 mol of Boc-protected aminooxy acetic acid; and for synthesis of the ketone peptide, weigh out 0.3 mol of 4-acetylbutyric acid. 4. Dissolve each compound (except for Fmoc-glycine) in 5 mL of DMF. For Fmoc-glycine, use 10 mL of DMF. 5. Load the reagents into the peptide synthesizer. Check the gas supply and other solutions to ensure that they are at the appropriate levels. 6. Follow the manufacturer’s instructions to synthesize the peptide in the following order: Resin-S-D-G-R-G-aminocaproic acid-(oxyamine or ketone) functional group. 7. Once the peptide synthesis program is completed, remove the resin with ethanol and place it inside a reaction vessel. 8. Remove the solvent by vacuum and add the cleaving solution to remove the peptide from the resin: Use 9.5 mL of TFA, 0.25 mL of water, and 0.25 mL of TIPS for the oxyamine; or use 9.0 mL of TFA 0.5 mL of water, and 0.5 mL of DCM for the ketone. Let the resin stir for 1 h. 9. Flush the TFA mixture into a round-bottom flask and then transfer the solution to a 50-mL centrifuge tube. 10. Precipitate the peptide with cold ethyl ether then spin down the peptide for 10 min at 3,000 × g. 11. Reprecipitate the peptide and spin down again at 3,000 × g.

Fabrication of Dynamic Self-Assembled Monolayers

429

12. Dissolve the peptide in water, and then freeze the solution against the side of a 50-mL centrifuge tube by rotating the centrifuge tube horizontally in dry ice. 13. Remove the cap from the centrifuge tube and secure a KimWipe with a rubber band to cover the open end. 14. Load the tube into the lyophilizer to dry overnight. Once dried, the final product should be a white solid with a structure corresponding to 2 (Fig.  1) for the oxyamine peptide and 3 (Fig. 1) for the ketone peptide. 3.1.7. Cell Culture

Cells must be split every few days in order to stay healthy. Perform the following procedures in a sterile laminar flow hood environment (see Note 6). 1. Remove a culture flask containing cells from the incubator (held at 37°C and 5% CO2) and place under a sterile laminar flow hood. Remove the growth medium by vacuum. 2. Gently wash the cells by pipetting 5  mL of PBS into the flask. 3. Remove the PBS by vacuum and add 1 mL of trypsinization solution to release the cells adhered to the bottom of the flask. Incubate the cells in trypsinization solution for 5 min. During this time, prepare a new culture flask with 5 mL of fresh serum-containing medium. 4. Add 5 mL of fresh serum-containing medium to the old flask containing the trypsinized suspension of cells and pipette up and down to remove the cells attached to the plastic. 5. Add a small amount of the diluted trypsinized cell suspension (<0.5  mL) to the new flask so that the cells have room to grow, and then place the newly seeded culture flask into the cell incubator. Transfer the remaining cells from the old culture flask to a 15-mL centrifuge tube.

3.1.8. Cell Seeding

Cells adhere to the surface when seeded and allow for the formation of cell patterns. 1. Take the 15-mL centrifuge tube containing the harvested cells (from Subheading 3.1.7, step 5) and place it in a centrifuge. Spin down the cells for 5 min at 1,000 × g. 2. Return to the sterile laminar flow hood and decant the serumcontaining medium from the centrifuge tube. 3. Resuspend the cells in 5 mL of serum-free media by pipeting 5  mL of DMEM up and down in the centrifuge tube (see Note 7). 4. If necessary, dilute the cells to 5,000 cells/mL in a new centrifuge tube with DMEM.

430

Westcott and Yousaf

5. Place the SAM surfaces with immobilized RGD peptides in a Petri dish and add enough cell solution to cover the SAM surface. 6. Place the cells in the CO2 incubator for 2 h so that the cells have time to attach to the SAM surface. 7. Remove the cells from the incubator and add fresh serumcontaining medium so that the medium completely immerses the SAM surfaces in the Petri dish. 3.1.9. Peptide Immobilization

Once the SAM surfaces are activated by photodeprotection (Subheading 3.1.5), they are incubated with a peptide solution. The resulting surface-immobilized peptides provide a ligand for cell attachment. 1. Place a small piece of Parafilm in an evaporation-proof Petri dish. 2. Add 60 mL of 10 mM peptide solution in water to the film. 3. Invert an activated SAM surface so that the gold side is touching the peptide solution. 4. Close the dish and incubate for 3 h. 5. Rinse the surface with ethanol.

3.2. Quinone Cell Patterning Strategy

The quinone strategy is based on the use of SAM surfaces containing hydroquinone-terminated thiols, which can be oxidized to quinones for ligand immobilization and cell patterning.

3.2.1. Chemical Surface Activation

Hydroquinone can be activated by an oxidant to generate the quinone. The quinone can then react with oxyamine-terminated peptides to generate a surface suitable for cell patterning. 1. To prepare photoprotected SAMs on gold-coated glass substrates, place the gold surfaces in 5 mL of a solution prepared by mixing 50 mL of a 1-mM photoprotected hydroquinone thiol (4, Fig. 1) solution and 4.95 mL of a 1-mM TEG (1) solution. Incubate the gold surfaces in the mixed thiol solution for 12 h (see Subheading 3.1.4) (see Note 8). 2. Deprotect the SAMs under a UV lamp (365 nm) following the general procedure described in Subheading 3.1.5. Only the photodeprotected areas of the substrates should now contain hydroquinone-terminated thiols (5, Fig. 1). 3. Place the photodeprotected SAM substrates into a saturated benzoquinone solution and wait for 1 h to convert the hydroquinone groups to the quinone form. 4. Rinse the activated surfaces with ethanol and immobilize RGD oxyamine peptides to the SAMs following the general procedure described in Subheading 3.1.9.

Fabrication of Dynamic Self-Assembled Monolayers

431

5. Seed the peptide-modified SAM substrates with cells following the general procedure described in Subheading 3.1.7. 3.2.2. Electrochemical Surface Activation

Hydroquinone can also be activated by applying a voltage to generate the quinone form. The quinone can then react with oxyamine-terminated peptides to generate a surface suitable for cell patterning. 1. Clip a photodeprotected SAM substrate (containing hydroxyquinone-terminated thiols) to the working electrode of a three-electrode setup having a Ag/AgCl reference electrode and a coiled platinum wire auxiliary electrode. 2. Perform cyclic voltammetry (CV) from −150 to 800 mV at 100  mV/s to verify the presence of the hydroquinone the surface. The oxidation peak occurs at 650 mV and the reduction occurs at 200 mV. (Fig. 3) 3. To activate the surface, perform linear sweep voltammetry (LSV) from −150 to 800 mV at 100 mV/s. 4. Rinse the surfaces with ethanol, and then immobilize RGD oxyamine peptides to the activated SAM following the general procedure described in Subheading 3.1.9.

3.2.3. Electrochemical Oxime Readout

Once an oxyamine ligand becomes immobilized to the activated SAM surface, the CV peaks shift to allow for identification of the covalent oxime conjugate product. 1. Reattach all electrodes to their corresponding leads after the oxyamine peptide immobilization step (Subheading 3.1.9) is completed.

Fig. 3. Cyclic voltammograms (CV) of hydroquinone and oxime SAMs. The front-most CV trace is for a hydroquinone SAM, with oxidation occurring at 650  mV and reduction occurring at 200  mV. After oxyamine immobilization to the surface, the redox peaks become shifted as seen in the background CV trace. The reduction peak shifts to 630 mV and the oxidation peak occurs at 400 mV.

432

Westcott and Yousaf

2. Perform a CV from −150 to 800 mV at 100 mV/s to verify the presence of the covalent oxime product at the surface. The oxidation peak shifts to 630 mV and the reduction shifts to 400 mV (Fig. 3). 3. Seed the peptide-modified SAM substrates with cells following the general procedure described in Subheading 3.1.7. 3.2.4. Cell Co-culture

By combining the hydroquinone surface chemistry strategy with microcontact printing, multiple cell lines can be patterned on a single surface (Fig. 4).

Fig. 4. Brightfield and fluorescent images of cell co-cultures grown on a photopatterned SAM substrate. Top panel: Brightfield image of cells patterned by photodeprotected HQ and microcontact printing. Cells in the top part of the image are 3T3-Swiss fibroblasts; cells at the bottom are transfected 3T3-Swiss albino fibroblast cells expressing green fluorescent protein (GFP)-actin. Bottom panel: Fluorescent image of the same cell coculture shown in the top panel. 3T3-Swiss albino cells in the top part of the image are nonfluorescent and do not appear visible in the fluorescent image, demonstrating that two distinct cell lines are present on the surface.

Fabrication of Dynamic Self-Assembled Monolayers

433

1. Soak a PDMS stamp (fabricated by following the general ­procedure described in Subheading 3.1.2) in a solution containing 1 mM hexadecane thiol in ethanol and dry with nitrogen (see Note 9). 2. Press the stamp against a clean gold surface and remove after 20 s. 3. Rinse the gold surface with ethanol. 4. Immerse the gold surface in 5 mL of a solution containing 50 mL of photoprotected hydroquinone thiol (4, Fig. 1) and 4.95 mL of TEG (1, Fig. 1). 5. Photodeprotect the SAMs on the gold surface by following the general procedure described in Subheading 3.1.5. Only the photodeprotected areas of the substrate should now contain hydroquinone-terminated thiols (5, Fig. 1). 6. Activate the hydroquinone groups either chemically (Subheading 3.2.1) or electrochemically (Subheading 3.2.2), as outlined above. 7. Seed the gold substrate with 3T3-Swiss albino fibroblasts expressing GFP-labeled actin following the general procedure described in Subheading 3.1.8. 8. Immobilize RGD oxyamine peptides to chemically activated regions of the gold substrate by immersing the substrate in a solution of 10 mM GRGDS–oxyamine in serum-containing media (see Subheading 3.1.9). 9. Seed the gold substrate with 3T3-Swiss albino fibroblasts following the general procedure described in Subheading 3.1.8 (Fig. 4). 3.3. Oxyamine Cell Patterning Strategy

The oxyamine strategy is based on the use of SAM surfaces containing photoprotected oxyamine thiols, which can be used to pattern cells. 1. To prepare photoprotected SAMs on gold-coated glass substrates, place the gold surfaces in 5 mL of a solution prepared by mixing 50 mL of a 1-mM photoprotected oxyamine thiol (6) solution and 4.95 mL of a 1-mM TEG (1, Fig. 1) solution. Incubate the gold surfaces in the mixed thiol solution for 12 h (see Subheading 3.1.4) (see Note 8). 2. Place a 60-mL drop of saturated semicarbizide solution between the photomask and the SAM surface. Deprotect the SAMs under a UV lamp (365 nm) following the general procedure described in Subheading 3.1.5. Only the photodeprotected areas of the substrates should now contain oxyamine-terminated thiols (7, Fig. 1).

434

Westcott and Yousaf

Fig. 5. Brightfield image of patterned 3T3-Swiss albino fibroblasts on photodeprotected oxyamine SAMs after ketone RGD immobilization.

3. Immobilize ketone-terminated GRGDS peptides to the SAMs  following the general procedure described in Subheading 3.1.9. 4. Seed the peptide-modified SAM substrates with cells following the general procedure described in Subheading  3.1.7 (Fig. 5). 3.4. Aldehyde Cell Patterning Strategy

The aldehyde strategy is based on the use of alcohol-terminated thiols, which are activated with a chemical oxidant to generate aldehydes that can be used to pattern cells. 1. To prepare SAMs on gold-coated glass substrates, place the gold surfaces in a solution containing 1 mM TEG (1, Fig. 1) in ethanol. Incubate the gold surfaces in the mixed thiol solution for 12 h (see Subheading 3.1.4). 2. Reversibly attach a microfluidic cassette with inlet/outlet access ports to the gold substrate by pressing it against the SAM surface to ensure a good seal (see Note 9). 3. Add 30 mL of a solution containing 60 mM PCC (8, Fig. 1) in acetonitrile to the inlet port on the microfluidic cassette. 4. Apply a vacuum to the outlet port on the microfluidic cassette to pump the PCC solution through the microfluidic channels. 5. Let the PCC solution sit inside the channels for 1 min. 6. Add 100 mL of ethanol to the inlet port on the microfluidic cassette, and then apply a vacuum to the outlet port to flush the ethanol through the microfluidic channels. 7. Remove the microfluidic cassette from the gold substrate and rinse the SAMs with ethanol. 8. Immobilize RGD oxyamine peptides to the SAMs following the general procedure described in Subheading 3.1.9. 9. Seed the peptide-modified SAM substrates with cells following the general procedure described in Subheading  3.1.7 (Fig. 6).

Fabrication of Dynamic Self-Assembled Monolayers

435

Fig. 6. Brightfield image of patterned cells on oxidized TEG SAMs after RGD oxyamine immobilization.

4. Notes 1. For the design of the photomasks used for fabricating the PDMS microfluidic cassettes, the areas corresponding to the flow channels will appear clear on the final photomask design while the background areas will be black. 2. For the design of the photomasks used for fabricating the PDMS microfluidic stamps, the background areas will appear clear in the final photomask design, whereas the desired shapes to be patterned will appear black. 3. To confirm that the bake time used for the photoresist is sufficient, remove the resist-coated wafer from the hot plate. If the resist layer begins to appear wrinkled, then more time is required to fully bake the photoresist on the hotplate. 4. After the photoresist is exposed and placed on a hotplate, the patterns should become visible in the postbaked photoresist. 5. Once gold has been evaporated onto glass substrates, the gold surfaces will oxidize within a few days. Therefore, it is best to use freshly prepared gold surfaces for cell biological experiments. 6. For cell work, make sure to thoroughly sterilize anything that is going to be placed inside the laminar flow hood with ethanol to reduce the possibility of contamination. 7. Cells are seeded onto surfaces in a serum-free medium in order to reduce the occurrence of nonspecific cell adhesion and migration.

436

Westcott and Yousaf

8. Once TEG SAMs have been formed on the gold substrates, store the surfaces in water to enhance their resistance to nonspecific protein and cell adhesion. 9. When not in use, the microfluidic stamps and cassettes should be stored in acetone. Before use, oven-dry the components to prevent fouling of the PDMS surfaces.

Acknowledgments This work was supported by the Carolina Center for Cancer Nanotechnology Excellence (NCI), and by grants from the Burroughs Wellcome Foundation (Interface Career Award) and the National Science Foundation (Career Award) to M.N.Y. References 1. Ridley, A. J., Schwratz, M. A., Burridge, K., Firtel, R. A., Ginsberg, M. H., Borisy, G., Parsons, J. T., and Horwitz, A. R. (2003) Cell migration: integrating signals from front to back. Science 302, 1704–1709. 2. Lutolf, M.P. and Hubell, J.A. (2005) Synthetic biomaterials as instructive 3. Extracellular microenvironments for morphogenesis in tissue engineering. Nat. Biomat. 23, 47–55. 4. Love, J. C., Estroff, L. A., Kriebel, J. K., Nuzzo, R. G., and Whitesides, G. M. (2005) Self-assembled monolayers of thiolates on metals as a form of nanotechnology. Chem. Reviews 105, 1103–1169. 5. Kane, R.S., Takayama, S., Ostuni, E., Ingber, D.E, and Whitesides, G.M. (1999) Patterning proteins and cells using soft lithography. Biomat. 20, 2363–2376. 6. Jiang, X., Bruzewicz, D.A., Wong, A.P., Piel,  M., and Whitesides, G.M. (2005)

Directing cell migration with asymmetric miropatterns. Proc. Acad. Natl. Sci. 102, 975–978. 7. Mahmud, G., Campbell, C.J., Bishop, K.J.M., Komarova, Y.A., Chaga, O., Soh, S., Huda, S., Kandere-Grzybowska, K., and Grzybowski, B.A. (2009) Directing cell motions on micropatterned ratchets. Nat Phys 5, 606–612. 8. Westcott, N.P., Pulsipher, A., Lamb, B.M. and Yousaf, M.N. (2008) Expedient generation of patterned surface aldehyde by Microfluidic oxidation for chemoselective immobilization of ligands and cells. Langmuir 24, 9237–9240 9. Park, S. and Yousaf, M.N. (2008) An interfacial oxime reaction to immobilize ligands and cells in patterns and gradients to photoactive surfaces. Langmuir 24, 6201–6207. 10. Lee, E., Chan, E.W.L., Yousaf, M.N. (2009) Spatio-temporal control of cell co-culture interactions on surfaces. ChemBioChem 8, 1648–1653.

Chapter 27 DNA Detection Using Functionalized Conducting Polymers Jadranka Travas-Sejdic, Hui Peng, Hsiao-hua Yu, and Shyh-Chyang Luo Abstract A well-defined DNA bioconjugated surface is a key component in the development of efficient biosensor platforms for diseases, ranging from point-of-care detection of pathogens and viruses to personalized diagnostics and medication, as well as for drug discovery, forensics, and food technology. We herein describe a universal and rapid methodology to construct such surfaces based on functionalized conducting polymer thin films. The conducting polymers combine sensing properties with the ability to act as signal transducers for the biorecognition event. We have shown that biosensor designs based on conducting polymers display a number of advantageous features, such as a long-term stability, label-free sensing, fast analysis, and the capability to apply both electrochemical and fluorescent protocols for DNA detection. Key words: Conducting polymer thin films, Biosensing, Polypyrroles, Polythiophenes, DNA, Electrochemical detection, Fluorescent detection

1. Introduction The completion of the human genome project provides new dimensions for clinical diagnostics and drug development based on an understanding of the role of specific DNA segments. As a result, there is a growing need for DNA-based technology platforms ranging from genotyping to molecular diagnostics. Among these new technologies, there is an increasing level of interest in the development of electrochemical DNA biosensors due to several factors, including (1) their cost-effectiveness coupled with modern semiconductor fabrication processes; (2) their high sensitivity with the possibility of further signal amplification through electrocatalysis (1–4); (3) their rapid and direct electrical detection (5–7) regardless of light-absorbing chemicals; and (4) the ease of manufacturing of portable, robust, low-cost, and easy-to-handle detection instrumentation suitable for field tests and home-care usage. Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_27, © Springer Science+Business Media, LLC 2011

437

438

Travas-Sejdic et al.

One of the key components for an electrochemical biosensor is efficient construction of a conductive biointerface. Construction of self-assembled monolayers (SAMs) based on the gold (Au)–thiol interaction is one of the most popular approaches for the production of a conductive biomaterial interface due to its simplicity, chemical availability, and the ability to produce thin, uniform films (8, 9). However, SAMs suffer from a number of disadvantages, including a limited selection of grafted electrode surfaces, the extended time needed for immobilization, and the potentialdependent instability of thiol–Au bonding (10–12). These drawbacks limit the development and large-scale manufacture of SAM-based electrochemical biosensors. Conducting polymers, on the other hand, are organic materials that have a unique electronic structure responsible for their electrical conductivity and interesting optical properties. Moreover, their electronic and optical proper­ ties are highly sensitive to the polymeric chain environment (13–16), which provides a means of generating a signal for binding a target analyte molecule (17). Utilizing conducting polymer films to develop novel DNA biosensors is a promising route due to the following advantages: (1) thin films can be electropolymerized onto electrodes of predetermined areas with a controlled thickness (<100 nm) (18–20), controlled roughness (Rrms < 5 nm) (18), and within a short period of time (usually within seconds); (2) they are conductive and therefore act as electrical signal transducers; (3) they are amenable to large-scale manufacturing; and (4) they exhibit very low intrinsic cytotoxicity (18, 21–23). Herein, we present general procedures for creating DNAbioconjugated conducting polymer interfaces for both optical and electrochemical DNA detection platforms. The methods described in this chapter have been developed in our two ­laboratories – at Polymer Electronics Research Centre (PERC) (Auckland, New Zealand) for the approaches based on functionalized polypyrroles (19, 20, 24, 25), and at RIKEN Advanced Science Institute (Japan) for the approaches based on functionalized poly(3,4-ethylenedioxythiophenes) (PEDOT) (26, 27). The methodologies used in our two laboratories follow the same basic procedure, but with a number of small variations.

2. Materials 2.1. Synthesis of Monomers 2.1.1. Synthesis of (3-Pyrrolyl)acrylic Acid

1. Pyrrole 98%, reagent grade, is distilled under vacuum before use. 2. Sodium hydride (NaH), 50% NaH dispersion in oil. 3. Triisopropylsilyl chloride, 97%.

DNA Detection Using Functionalized Conducting Polymers

439

4. N-Bromosuccinimide, ³95%. 5. n-Butyllithium, 1.6 M in hexane. 6. (Triphenylphosphoranylidene) methyl acetate, 98%. 7. Dimethylformamide (DMF), anhydrous. 8. Sodium hydroxide (NaOH), 10% (w/v) aqueous solution. 9. Solvents: hexane, ethyl acetate, ether, and methanol. 10. Hydrochloric acid (HCl), 10% (w/v) aqueous solution. 11. Drying reagents: sodium sulfate and magnesium sulfate. 12. Silica gel for column chromatography. 13. Neutral alumina. 14. 1H and 13C nuclear magnetic resonance (NMR) spectra were acquired at room temperature using a 300-MHz spectrometer (Bruker). 15. Mass spectrometry (MS) instrumentation. 2.1.2. Synthesis of Carboxylic Acid-Functionalized 3,4-Ethylenedioxythio­ phene

1. Hydroxymethyl-functionalized 3,4-ethylenedioxythiophene (EDOT+OH). This reagent may either be purchased from commercial suppliers or synthesized according to the literature procedure (28). 2. Tetrahydrofuran (THF), anhydrous in a sure-seal bottle. 3. Sodium iodide (NaI). 4. Sodium hydride (NaH), 60% (w/w) suspension in mineral oil. 5. Methyl bromoacetate. 6. HCl solution, 1N. 7. Argon. 8. Ethyl acetate. 9. Sodium hydroxide (NaOH). 10. Magnesium sulfate, used as received. 11. Standard Schlenk lines and glassware. 12. CombiFlash® Companion® flash chromatography system on normal phase silica gel cartridges (ISCO). 13. 1H and 13C NMR data were acquired at 25°C with an AV 400 Spectrometer (Bruker). 14. Finnigan MAT LCQ Mass Spectrometer (Thermo Finnigan, San Jose, CA) fitted with an electrospray ionization (ESI) probe.

440

Travas-Sejdic et al.

2.2. Electrochemical Polymerization of Conducting Polymer Sensing Films 2.2.1. Electrochemical Copolymerization of (3-Pyrrolyl)acrylic Acid and Pyrrole

1. Working electrode: glassy carbon electrode (3.0  mm i.d.) (Bioanalytical Systems, Inc., West Lafayette, IN, No. MF-2012). 2. Reference electrode: silver/silver chloride (Ag/AgCl) (BASi, MF-2052). 3. Auxiliary electrode: platinum (Pt) wire (BASi, No. MW-1032), used as received. 4. Alumina slurry (0.5 mm) (Allied Tech Products, Inc, USA). 5. Cell vials (2–15 mL) (BASi, No. MF-1084). 6. Electrocopolymerization solution: 0.5 M Pyrrole, 6.25 mM (3-Pyrrolyl)acrylic Acid (PAA), 0.2 M LiClO4 in aceto­nitrile. 7. CH Instruments electrochemical workstation, Model 440 (CH Instruments, USA).

2.2.2. Electrochemical Copolymerization of Hydroxyl-Functionalized and Carboxylic-Acid Functionalized 3, 4-Ethylenedioxythiophene

1. Working electrodes: indium tin oxide (ITO)-coated glass plates (Delta Technologies, Ltd., Stillwater, MN), gold (Au) and platinum (Pt) disk (BASi., No. 002421, 002422). 2. Reference electrode: silver/silver chloride (Ag/AgCl), (BASi, No. RE-1C) used as received. 3. Auxiliary electrode: Pt gauze and Pt wire counter electrode (Aldrich, No. 444685, 298093), used as received. 4. Polishing Kit PK-4 (Bioanalytical Systems, Inc.) and alumina slurry (0.05-mm, Micropolish Gamma) (Buehler, Lake Bluff, IL). 5. Monomer solution: aqueous solution containing 0.01  M total EDOT monomer, 0.1 M lithium perchlorate (LiClO4), 0.1 M hydrochloric acid (HCl), and 0.05 M sodium dodecyl sulfate (SDS).

2.3. Immobilization of Oligonucleotide Probes

1. The immobilization of amino-oligonucleotide (ODN) onto the carboxylic functionalized polymer films is performed using a carbodiimide catalyst either with (Subheading 2.3.1) or without N-hydroxysulfosuccinimide (NHS) activating agent (Subheading 2.3.2) (see Note 1).

2.3.1. Immobilization of ODN Probes onto Functionalized Polypyrrole Films

1. Immobilization agent: 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDC), ³97%.

2.3.2. Immobilization of ODN Probes onto Functionalized PEDOT Films

1. Activation/immobilization reagents: N-Hydroxysulfosuc­ cinimide (sulfo-NHS) (Pierce) and 1-ethyl-3-[3-dimethylaminopropyl] carbodiimide hydrochloride (EDC) (Pierce) solution.

2. Immobilization buffer: 10  mM phosphate-buffered saline (PBS), pH 5.2.

2. Immobilization buffer: PBS consisting of 137  mM NaCl, 2.7 mM KCl, and 10 mM phosphate buffer, pH 7.4.

DNA Detection Using Functionalized Conducting Polymers

441

2.4. ODN Biosensing

1. Hybridization buffer: PBS, pH 7.4 (40 mL).

2.4.1. ODN Biosensing with Functionalized Polypyrrole Films

2. Synthetic, single-stranded, 18-mer ODNs with complementary, noncomplementary sequences (Invitrogen, Carlsbad, CA; or Alpha DNA, Montreal, Canada). 3. PBS (pH 7.4) containing 5.0  mM K4Fe(CN)6/K3Fe(CN)6 (1:1 mol/mol). 4. AC impedance measurement: EG&G potentiostat/galvanostat (Model 280, Princeton Applied Research) with EG&G 1025 Frequency Response Analyzer; cell vials (as in Subheading 2.2.1, item 5); Ag/AgCl reference electrodes (as described in Subheading 2.2.1); Pt sheet (1.5 × 1.5 cm). 5. Data analysis: ZView software (Version 2.80) (Scribner Associates, Inc., Southern Pines, NC).

2.4.2. ODN Biosensing with Functionalized PEDOT Films

1. Electrochemical reporting enzyme solution: glucose oxidaseavidin D (GOD-A) (Vector Laboratories), diluted 1:100 (v/v) in PBS (pH 7.4) to form a 50-mg/mL working solution. 2. Redox polymer: poly(vinylimidazole)-polymer(acrylamide) copolymer partially imidazole-complexed with Os(4,4¢dimethyl-2,2¢-bipyridine)2Cl+/2+ (PVA-Os), synthesized as described in ref. (29). 3. Cy3-labeled complementary Carlsbad, CA).

DNA

target

(Invitrogen,

3. Methods 3.1. Synthesis of Carboxylic Acid Group-Functionalized Pyrrole: PAA 3.1.1. Synthesis of N-(Triisopropylsilyl) pyrrole (1)

The synthetic scheme of PAA is shown in Fig. 1.

1. Pyrrole (1.0 mL, 0.96 g, 15 mmol) is added dropwise at 0°C to a stirred suspension of sodium hydride (0.758  g of 50% dispersion in mineral oil, 16  mmol) in anhydrous DMF (20 mL) in a 50-mL round-bottom flask. 2. After hydrogen evolution (foaming) ceases, triisopropylsilyl chloride (3.1 mL, 2.8 g, 15 mmol) is added dropwise, and the solution is stirred continuously at 0°C for 45 min. 3. The reaction mixture is partitioned between ether and water, and the ether phase is washed with water, dried over sodium sulfate, and evaporated with a rotary evaporator under reduced pressure.

442

Travas-Sejdic et al. O C

Br

N

N

N

Si

Si

Si

1

2

H

3 O C

OCH3

O C

OH N Si

N H 5

4

Fig. 1. Schematic representation of the synthesis of (3-pyrrolyl)acrylic acid (PAA).

4. The product is purified by column chromatography on silica gel using hexane/ethyl acetate = 10:1 (by volume) as an eluent to give an oil (1.01 g, 30.2% yield). 5. The compound 1 is characterized by 1H NMR (300 MHz, CDCl3): d 1.09 (d, 18 H), 1.45 (sept, 3 H), 6.31 (t, 2 H), 6.80 (t, 2 H) and mass spectroscopy (M + H) m/z: 224. 3.1.2. Synthesis of 3-Bromo-1(Triisopropylsilyl)pyrrole (2)

1. N-Bromosuccinimide (NBS, 0.8 g, 4.5 mmol) is added to a stirred solution of N-(triisopropylsily1)pyrrole (1.0  g, 4.5 mmol) in anhydrous THF (10 mL) at −78°C in a dry ice/ acetone bath. The reaction mixture is kept at −78°C for 1–2 h and then left to cool down to room temperature (ca. 1 h). 2. Pyridine (0.1 mL) and hexane (10 mL) are added to the reaction mixture and the resulting suspension is filtered through a plug of neutral alumina. 3. The filtrate is evaporated in vacuum. Purification by column chromatography on silica gel using hexane/ethyl acetate = 10:1 (by volume) as an eluent gives an oil (1.22  g, 90.1% yield). 4. The purity of the compound 2 is checked by 1H NMR (300  MHz, CDCl3): d 1.08 (d, 18 H), 1.42 (sept, 3 H), 6.26 (dd, 1H), 6.66 (t, 1H), 6.72 (dd, 1H) and MS (M + H) m/z: 302.

DNA Detection Using Functionalized Conducting Polymers 3.1.3. Synthesis of l-(Triisopropylsilyl) pyrrole-3Carboxaldehyde (3)

443

1. A solution of 1.6  M n-butyllithium in hexane (1.56  mL, 2.5 mmol) is added, at −78°C in a dry ice/acetone bath, to a solution of 3-bromo-l-(triisopropylsilyl)pyrrole (0.76  g, 2.5 mmol) in anhydrous THF (20 mL). 2. After 15 min at −78°C, DMF (2.5 mmol) is added; 15 min thereafter, the reaction is removed from the cooling bath and left to warm up to room temperature. 3. The reaction mixture is quenched with 20 mL of water and extracted three times with 10 mL of ether. Following this, the extract is dried over MgSO4 and evaporated with a rotary evaporator under reduced pressure. 4. The residue is then purified by column chromatography on silica gel using hexane/ethyl acetate (7:3, v/v) as an eluent to give an oil (0.501 g, yield 80%). 5. The purity of the compound 3 is checked by 1H NMR (300 MHz, CDCl3): d 1.10 (d, 18 H), 1.47 (sept, 3 H), 6.74 (m, 1H), 6.77 (m, 1H), 7.39 (dd, 1H), 9.82 (m, 1H) and MS (M + H) m/z: 252.

3.1.4. Synthesis of 3-(l-(Triisopropylsilyl) pyrrol-3-yl)-acrylic Acid Methyl Ester (4)

1. (Triphenylphosphoranylidene) methyl acetate (3  mmol, 1.05 g, 1.5 equiv) is added to a solution of l-(triisopropylsilyl) pyrrole-3-carboxaldehyde (0.502 g, 2.0 mmol) in anhydrous THF (20 mL). 2. The reaction mixture is stirred at 30°C for 2 h and then concentrated with a rotary evaporator under reduced pressure to give an oily residue, which upon purification by column chromatography on silica gel using hexane/ethyl acetate (10:1, v/v) as an eluent gives the product 4. 3. The purity of the compound 4 is checked by 1H NMR (300 MHz, CDCl3): d 1.04 (d, 18 H), 1.45 (sept, 3 H), 3.74 (s, 3 H), 6.10 (d, 1H ), 6.50 (m, 1H ), 6.74 (m, 1H), 6.97 (m, 1H), 7.69 (d, 1H) and MS (M + H) m/z: 308.

3.1.5. Synthesis of PAA (5)

1. 3-(l-(Triisopropylsilyl)pyrrol-3-yl)-acrylic acid methyl ester (3 mmol, 0.924 g) is added to an aqueous solution of 10 wt% NaOH (6  mL) and methanol (3  mL), and the mixture is refluxed for 2 h. Methanol is then removed by distillation in vacuum. 2. The aqueous solution is adjusted to pH 4.3 with 10% (w/v) HCl. The resultant precipitate is collected, washed with cold water, and recrystallized from an ethanol–hexane mixture to give 5. 3. The purity of the compound 5 is checked by 1H NMR (300 MHz, (CH3)SO-d6): d 5.95 (d,1H), 6.40 (m, 1H), 6.80 (m, 1H), 7.18 (m, 1H), 7.46 (d, 1H), 11.15 (s, 1H), 11.80 (s, 1H).

444

Travas-Sejdic et al. H3CO

HO O

O

OH

O

O

O

S EDOT-OH

O

O

O

S EDOT-COOHCH3

O

O

S EDOT-COOH

Fig.  2. Schematic representation of the synthesis of carboxylic acid-functionalized 3,4-ethylenedioxythiophene (EDOT-COOH).

3.2. Synthesis of Carboxylic Acid-Functionalized 3,4-Ethylenedioxy­ thiophene 3.2.1. Synthesis of Methyl Carboxylate-Functionalized 3,4-Ethylenedioxythiophene

The detailed synthesis scheme is shown in Fig. 2.

1. A 100-mL round-bottom flask is charged with a stir bar and the reaction mixture, consisting of EDOT-OH (861  mg, 5.0 mmol), NaI (150 mg, 1.0 mmol), and NaH (60% w/w suspension in mineral oil, 240  mg, 6.0  mmol). The flask is evacuated in vacuum and then filled with argon using a dual Schlenk line, with one line connected to vacuum and the other line connected to the argon supply. This procedure needs to be repeated at least three times (see Note 2). 2. Anhydrous THF (stored in an inert gas-filled bottle) is drawn by syringe (20 mL) and introduced into the reaction mixture. After THF is introduced, the suspension is stirred for 15 min and cooled in an ice bath. 3. Methyl bromoacetate (0.57 mL, 0.92 g, 6.0 mmol) is then added dropwise, and the reaction mixture is stirred for 18 h (see Note 3). 4. The majority of THF is removed with a rotary evaporator under reduced pressure; the crude product is partitioned between water and ethyl acetate, and the aqueous layer is extracted with 100 mL of ethyl acetate twice. 5. The combined organic layers are washed with saturated sodium chloride solution. The solution is then dried with anhydrous magnesium sulfate, and the magnesium sulfate drying reagent

DNA Detection Using Functionalized Conducting Polymers

445

is filtered. The filtrate is then evaporated to remove the organic solvent, and the crude product is thereby obtained. 6. The crude product is further purified with a silica gel column using hexane/ethyl acetate (5:1, v/v) as an eluent to give the purified intermediate methyl carboxylate-functionalized 3,4-ethylenedioxythiophene (EDOT-COOCH3) (see Note 4). 7. The purity of the compound EDOT+COOCH3 is checked by 1 H NMR (400 MHz, CDCl3): 6.35 (d, 1H, J = 3.6 Hz), 6.33 (d, 1H, J = 3.6 Hz), 4.37 (ddd, 1H, J = 11.6, 7.6, 2.0 Hz), 4.28 (dd, 1H, J = 11.6, 2.0 Hz), 4.19 (s, 2 H), 4.12 (dd, 1H, J = 11.6, 7.6 Hz), 3.84 (dd, 1H, J = 10.4, 5.2 Hz), 3.78 (dd, 1H, J = 10.4, 5.2 Hz), 3.76 (s, 3 H) and 13C NMR (100 MHz, CDCl3): 171.1, 142.0, 141.9, 100.5, 100.4, 73.3, 70.5, 69.5, 69.3, 52.6. Analytically calculated for C10H12O5S: C, 49.17; H, 4.95; found: C, 48.87; H, 4.70. 3.2.2. Synthesis of Carboxylic Acid-Functionalized 3,4-Ethylenedioxythio­ phene

1. The product EDOT+COOCH3 (610 mg, 2.5 mmol) is dissolved in THF (10 mL) in a 100-mL round-bottom flask. 2. Freshly prepared aqueous NaOH solution (2 M, 10 mL) is added to the reaction mixture, and the mixture is stirred vigorously until the starting material is completely consumed (typically about 3  h). The hydrolysis is monitored by thin layer chromatography (TLC). 3. The mixture is then acidified to pH < 3 by the addition of 1N HCl solution. The reaction mixture is then extracted with ethyl acetate five times and the organic phases are combined (see Note 5). 4. The combined organic layers are washed with water until a neutral pH is obtained. The solution is then dried with anhydrous MgSO4, and the MgSO4 is filtered. 5. The filtrate is then evaporated to remove the organic solvent. The product thereby obtained, carboxylic acid-functionalized 3,4-ethylenedioxythiophene (EDOT-COOH) (480 mg, 83%), is a thick colorless oil that solidifies upon standing overnight. 6. The purity of the compound EDOT-COOH is checked by 1 H NMR (400 MHz, CDCl3): 6.36 (d, 1H, J = 3.6 Hz), 6.34 (d, 1H, J = 3.6 Hz), 4.38 (ddd, 1H, J = 11.6, 7.2, 2.4 Hz), 4.26 (dd, 1H, J = 11.6, 2.4 Hz), 4.24 (s, 2 H), 4.12 (dd, 1H, J = 11.6, 7.2 Hz), 3.85 (dd, 1H, J = 10.4, 4.8 Hz), 3.81 (dd, 1H, J = 10.4, 4.8  Hz) and 13C NMR (150  MHz, CDCl3): 175.2, 141.3, 141.1, 100.0, 99.9, 72.6, 70.0, 68.4, 65.8. HR-MS (FAB): [M + H] calculated for C9H11O5S: 231.0237; found: 231.0237.

446

Travas-Sejdic et al.

3.3. Electrochemical Polymerization of Conducting Polymer Sensing Films 3.3.1. Electrochemical Copolymerization of PAA and Pyrrole

1. A polymerization solution containing 0.5 M pyrrole, 6.25 mM PAA, and 0.2  M LiClO4 in 2  mL of acetonitrile is charged into a three-electrode cell setup comprising of a precleaned glassy carbon working electrode (3.0 mm i.d., 7.07 mm2 geometric surface area) (see Note 6), an Ag/AgCl (3  M KCl) reference electrode and a Pt wire counter electrode. 2. The electrochemical copolymerization of PAA with pyrrole is carried out using a CH Instruments electrochemical workstation (Model 440) at a constant potential of 1.0 V (vs. Ag/ AgCl) until a total charge of 2.12  mC/cm2 is reached that approximates to a 10 nm film thickness (see Note 7). 3. After polymerization, the film on the electrode is washed with acetonitrile and PBS buffer (pH 7.4), and then stored in PBS, pH 7.4.

3.3.2. Electrochemical Copolymerization of EDOT–COOH

1. A three-electrode cell with a precleaned ITO-coated glass electrode (see Note 8) as the working electrode, a silver/silver chloride (Ag/AgCl) reference electrode, and Pt wire or gauze as an auxiliary electrode is set up. The polymerization solution contains 10 mM total EDOT monomer concentration, with 0.1–1% EDOT-COOH and 99.9–99% EDOT-OH, and other electrolytes as described in Subheading 2.2.2, item 5. 2. The polymer films are electrochemically copolymerized onto the working electrode by applying a constant potential of 1.0 V until a total charge of 60 mC/cm2 is reached, at which point the film is approximately 30-nm in thickness. 3. The polymer films are rinsed with deionized (DI) water two times before oligonucleotide probe conjugation (see Subheading 3.4).

3.4. Immobilization of Oligonucleotide Probes 3.4.1. Immobilization of ODN Probes onto Carboxylic Acid-Functionalized Polypyrrole Films 3.4.2. Immobilization of ODN Probes onto Carboxylic AcidFunctionalized PEDOT Films

1. 40 ml of phosphate buffer (pH 5.2) containing 20 nmol of amine-modified ODN probe and 400  nmol of EDC is injected onto the surface of the copolymer-coated electrode (see Note 9). 2. The electrode is kept at 28°C for 1 h. 3. The modified electrode is thoroughly washed using PBS (pH 7.4) in order to remove any unattached ODN probes, and blown dry with nitrogen. 1. A solution containing 20  mM EDC and 80  mM N-hydroxysulfosuccinimide (sulfo-NHS) in deionized water is mixed, vortexed for 10 s, and then immediately drop-cast onto the conducting polymer films. 2. After incubation at room temperature for 15–20 min, excess sulfo-NHS and EDC are rinsed and removed by washing with deionized water.

DNA Detection Using Functionalized Conducting Polymers

447

3. Amine-modified ODN probes are immobilized onto the films by immersing the activated conducting polymer films in PBS buffer (pH 7.4) containing 1 mM ODN for 5 h. 4. After immobilization, the conducting polymer films are rinsed with PBS (pH 7.4) and water two times, and then blown dry with nitrogen to remove nonspecifically adsorbed DNA probes. 3.5. ODN Biosensing

3.5.1. Electrochemical ODN Detection with ODN-Conjugated Polypyrrole Films

In this section, we demonstrate (1) the direct, label-free electrochemical detection of target ODNs complementary to the probe ODN using ODN-conjugated polypyrrole films and (2) both fluorescent- and enzyme-catalyzed electrochemical-based target ODN detection using ODN-conjugated PEDOT films. 1. Hybridization of target ODNs is performed by injecting 40 mL of PBS solution (pH 7.4) containing ODN target onto the surface of the ODN-probe conjugated polymer film and incubating the film at 42°C in a moisture-saturated chamber for 1  h. The film is then washed three times with PBS to remove any nonybridized ODNs. 2. Electrochemical impedance measurements are undertaken using a conventional three-electrode cell comprised of a polymer film-modified glassy carbon electrode as working electrode, Pt foil as a counter electrode, and an Ag/AgCl (in saturated KCl) reference electrode. 3. Electrochemical impedance spectra (EIS) are measured in 10  mL of PBS solution (pH 7.4) containing 5.0  mM K4Fe(CN)6/K3Fe(CN)6 (1:1 mol/mol) as the redox couple. Prior to taking measurements, the solution is purged with nitrogen for 5 min. 4. EIS measurements are taken before ODN probe immobilization, after ODN probe immobilization, and after hybridization with the target ODNs. EIS are measured at an applied bias potential of +230 mV at 20°C, using the [Fe(CN)6]3−/4− redox couple, and with a 10-mV sinusoidal excitation amplitude over the frequency range of 1 Hz to 2 MHz at 12 steps per decade by means of an EG&G potentiostat/galvanostat coupled with an EG&G 1025 Frequency Response Analyzer. 5. The experimental data are analyzed by using ZView software. 6. A modified Randles equivalent circuit (Fig. 3) consisting of a solution resistance (Rs) in series with a constant phase element (CPE) parallel with a charge-transfer resistance (Rct) and Warburg impedance (W) is used to model the experimental data. 7. A typical result is shown in Fig. 4, which presents the EIS of a polymer film-coated electrode before and after immobilization

448

Travas-Sejdic et al.

Fig. 3. Equivalent circuit model consisting of a solution resistance, Rs, a charge transfer resistance, Rct, a Warburg impedance, W, and a constant phase element, CPE. 200

Zim (ohm)

160 120 80

c

40 0

b a

100

200

300

400

500

600

700

Zre (ohm)

Fig. 4. Electrochemical impedance spectra (presented as Nyquist plots (−Zim vs. Zre)) for an electrode coated with poly(Py-co-PAA) in PBS solution (pH 7.4) containing 5.0 mM Fe(CN)63−/Fe(CN)64−: (a) before immobilization of ODN probes; (b) after immobilization of ODN probes; and (c) after hybridization with 20.2  nM of complementary ODNs. Reproduced with permission from ref. 25 © 2007 Elsevier.

Table 1 Values of the fitted equivalent circuit parameters derived for the curves shown in Fig. 4 Rs (W)

Rct (W)

CPE (mF)

n

c2

Before probe immobilization

57.8 ± 0.4

55.0 ± 0.8

5.04 ± 0.38

0.77 ± 0.01

2.31 × 10−4

After probe immobilization

59.5 ± 0.4

173.8 ± 2.7

4.89 ± 0.25

0.77 ± 0.007

4.02 × 10−4

After hybridization

57.6 ± 0.4

225 ± 6.3

4.21 ± 0.64

0.72 ± 0.008

5.4 × 10−4

Reproduced with permission from Ref. (25) © 2007 Elsevier

of ODN probes, and after hybridization with 20.2 nM of fully complementary ODN target. The fitted values for the parameters of the equivalent circuit (Fig. 3) are given in Table 1. The very low chi-squared values suggest that this model fits the experimental data well. After immobilization of the ODN probes, the value of Rct increased from 55 to 174 W; and after hybridization with 20.2 nM of fully complementary ODN target, Rct increased significantly further to 255 W.

DNA Detection Using Functionalized Conducting Polymers

449

Fig. 5. Fluorescence intensity levels measured on a ODN-bioconjugated PEDOT conducting polymer biointerface in the presence of 1 mM Cy3-labeled complementary target DNA (black ), one-base mismatch target DNA (gray ), and noncomplementary target DNA (white) in fluorescence labeling experiments. The mol% of carboxylic acid-functionalized 3,4-ethylenedioxythiophene (EDOT-COOH) monomer polymerized on different indium tin oxide (ITO) substrates is indicated. Reproduced with permission from ref. 27 © 2009 American Chemical Society. 3.5.2. Fluorescent and Electrochemical ODN Detection with ODN-Conjugated PEDOT Films 3.5.2.1. Demonstration of Fluorescence-Based Detection Using Cy3-Labeled Target DNA

3.5.2.2. Demonstration of Electrochemical Detection Using Biotin-Labeled Target ODNs

1. PEDOT–ODN immobilized films are immersed in a PBS solution (pH 7.4) containing 1 mM Cy3-labeled target DNA and incubated overnight in a moisture-saturated chamber maintained at room temperature. 2. After hybridization of the target ODNs, the polymer films are washed with PBS (pH 7.4) three times and blown dry. The polymer films are stored in the dark until fluorescence measurements are taken. 3. Fluorescence measurements are performed with a fluorescence microscope (BX-51, Olympus) with a CCD camera (DP70, Olympus). Images are then analyzed with Image-Pro 3D Suite software (Media Cybernetics) to calculate the fluorescence intensity. The summarized results are shown in Fig. 5. 1. The conducting polymer sensing films are immersed in a PBS solution (pH 7.4) containing 1  mM biotin-labeled target ODNs overnight. 2. After rinsing thoroughly with PBS (pH 7.4) and deionized water two times, the electrodes (containing hybridized biotin-labeled target ODNs) are immersed in GOD-A solution for 30 min. 3. The substrates are washed with PBS (pH 7.4) three times and kept in PBS solution before electrochemical detection. 4. The glucose electrooxidation current is measured amperometrically in a PBS solution (pH 7.4) containing 40 mM glucose at a potential of 0.4 V (vs. Ag/AgCl) in a Faraday cage. The summarized results are shown in Fig. 6.

450

Travas-Sejdic et al.

Fig. 6. (a) In situ amperometric response of glucose oxidation on poly(EDOT-OH)-copoly(EDOT-COOH) films (prepared from 0.2 mol% of C2-EDOT-COOH solutions) after hybridization with complementary target DNA at different concentrations. (b) Dynamic range and detection limit for the PEDOT-based electrochemical platform. Reproduced with permission from ref. (27) © 2009 American Chemical Society.

4. Notes 1. The NHS route may provide more efficient coupling of amine-ODNs onto carboxylic acid-functionalized conducting polymer films. 2. The weighing paper or container for NaH must be quenched by methanol or acetone before being thrown away. 3. Larger scale synthesis usually takes longer than 24 h. 4. Because the end product is difficult to purify by chromatography, the purification of this intermediate product is crucial

DNA Detection Using Functionalized Conducting Polymers

451

to the purity of the end product. This step must be handled carefully. 5. The acidic solution may cause polymerization of monomers. Therefore, the pH of the solution must be carefully controlled. 6. The glassy carbon electrode is polished with a 0.5-mm alumina slurry, then sonicated in 5 mL of acetonitrile containing 0.2 g of active carbon powder, then in 5 mL of chloroform and 10 mL of Milli-Q water for 10 min, respectively. 7. The thickness of the film is controlled by the total charge passed during polymerization. The relationship between the film thickness and the charge is as follows: Film thickness (nm) = 4.76 × C (mC)/electrode area (cm2) (30). 8. Indium tin oxide (ITO)-coated glass plates are cleaned by immersion in a detergent solution, acetone, dichloromethane, and finally in isopropanol with ultrasonic agitation for a period of 30  min prior to use. Gold (Au) and platinum (Pt) disk working electrodes are polished by using a PK-4 Polishing Kit (BASi) and with 0.05-mm alumina (Micropolish Gamma) (Buehler) before use. 9. EDC solutions should be freshly prepared and used only once.

Acknowledgments J.T.S. and H.P. thank the Royal Society New Zealand (Marsden Fund) and the Auckland UniServices for financial support. The work performed by H.H.Y. and S.C.L. has been supported by the Institute of Bioengineering and Nanotechnology (Biomedical Research Council, Agency for Science, Technology and Research, Singapore) and the RIKEN Advanced Science Institute (Japan). References 1. Gooding, J. J. (2002) Electrochemical DNA hybridization biosensors. Electroanalysis 14, 1149–1156. 2. Christopoulos, T. K. (1999) Nucleic acid analysis. Anal. Chem. 71, 425R-438R. 3. Wang, J. (2002) Electrochemical nucleic acid biosensors. Anal. Chim. Acta 469, 63–71. 4. Wang, J. (2005) Carbon-nanotube based electrochemical biosensors: A review. Electroanalysis 17, 7–14. 5. Wang, J. (2000) From DNA biosensors to gene chips. Nucleic Acids Res. 28, 3011–3016.

6. Sassolas, A. Leca-Bouvier, B. D., and Blum, L. J. (2008) DNA biosensors and microarrays. Chem. Rev. 108, 109–139. 7. Katz, E. and Willner, I. (2003) Probing biomolecular interactions at conductive and semiconductive surfaces by impedance spectroscopy: Routes to impedimetric immunosensors, DNASensors, and enzyme biosensors. Electroanalysis 15, 913–947. 8. Kjallman, T. H. M., Peng, H., Soeller, C., and Travas-Sejdic, J. (2008) Effect of probe density and hybridization temperature on the

452

Travas-Sejdic et al.

response of an electrochemical hairpin-DNA sensor. Anal. Chem. 80, 9460–9466. 9. Fan, C., Plaxco, K. W., and Heeger, A. J. (2003) Electrochemical interrogation of conformational changes as a reagentless method for the sequence-specific detection of DNA. Proc. Natl. Acad. Sci. USA 100, 9134–9137. 10. Everett, W. R., Welch, T. L., Reed, L., and Fritschfaules, I. (1995) Potential-dependent stability of self-assembled organothiols on gold electrodes in methylene-chloride. Anal. Chem. 67, 292–298. 11. Beulen, M. W. J., Kastenberg, M. I., van Veggel, F., and Reinhoudt, D. N. (1998) Electrochemical stability of self-assembled monolayers on gold. Langmuir 14, 7463–7467. 12. Zhong, C. J. and Porter, M. D. (1997) Fine structure in the voltammetric desorption curves of alkanethiolate monolayers chemisorbed at gold. J. Electroanal. Chem. 425, 147–153. 13. Kirchmeyer, S., and Reuter, K. (2005) Scientific importance, properties and growing applications of poly( 3,4-ethylenedioxythiophene). J. Mater. Chem. 15, 2077–2088. 14. Groenendaal, L., Zotti, G., Aubert, P. H., Waybright, S. M., and Reynolds, J. R. (2003) Electrochemistry of poly(3,4-alkylenedioxythiophene) derivatives. Adv. Mater. 15, 855–879. 15. Groenendaal, B. L., Jonas, F., Freitag, D., Pielartzik, H. and Reynolds, J. R. (2000) Poly(3,4-ethylenedioxythiophene) and its deri­ vatives: Past, present, and future. Adv. Mater. 12, 481–494. 16. Pei, Q. B., Zuccarello, G., Ahlskog, M., and Inganas, O. (1994) Electrochromic and highly stable poly(3,4-ethylenedioxythiophene) switches between opaque blue-black and transparent sky blue. Polymer 35, 1347–1351. 17. Travas-Sejdic, J. and Soeller, C. (2008) Sensing genes using conducting polymers, nanoparticles, and nanotubes. In Handbook of Organic Electronics and Photonics, Volume 1 (Nalwa, H.S., ed.), American Scientific Publishers, Valencia, CA, pp. 365–403. 18. Luo, S. C., Ali, E. M., Tansil, N. C., Yu, H. H., Gao, S., Kantchev, E. A. B., and Ying, J. Y. (2008) Poly(3,4-ethylenedioxythiophene) (PEDOT) nanobiointerfaces: Thin, ultrasmooth, and functionalized PEDOT films with in vitro and in vivo biocompatibility. Langmuir 24, 8071–8077. 19. Peng, H., Soeller, C., Cannell, M. B., Bowmaker, G. A., Cooney, R. P., and TravasSejdic, J. (2006) Electrochemical detection of DNA hybridization amplified by nanoparticles. Biosens. Bioelectron. 21, 1727–1736.

20. Peng, H.; Soeller, C., Vigar, N., Kilmartin, P. A., Cannell, M. B., Bowmaker, G. A., Cooney, R. P. and Travas-Sejdic, J. (2005) Label-free electrochemical DNA sensor based on functionalized conducting copolymer. Biosens. Bioelectron. 20, 1821–1828. 21. Martin, D. C. (2007) Organic electronics: Polymers manipulate cells. Nat. Mater. 6, 626–627. 22. Isaksson, J., Kjall, P., Nilsson, D., Robinson, N. D., Berggren, M., and Richter-Dahlfors, A. (2007) Electronic control of Ca2+ signaling in neuronal cells using an organic electronic ion pump. Nat. Mater. 6, 673–679. 23. Richardson-Burns, S. M., Hendricks, J. L., Foster, B., Povlich, L. K., Kim, D. H., and Martin, D. C. (2007) Polymerization of the conducting polymer poly(3,4-ethylenedioxythiophene) (PEDOT) around living neural cells. Biomaterials 28, 1539–1552. 24. Peng, H., Soeller, C., and Travas-Sejdic, J. (2007) Novel conducting polymers for DNA sensing. Macromolecules 40, 909–914. 25. Peng, H., Soeller, C., Vigar, N. A., Caprio, V., and Travas-Sejdic, J. (2007) Label-free detection of DNA hybridization based on a novel functionalized conducting polymer. Biosens. Bioelectron 22, 1868–1873. 26. Ali, E. M., Kantchev, E. A. B., Yu, H. H., and Ying, J. Y., (2007) Conductivity shift of polyethylenedioxythiophenes in aqueous solutions from side-chain charge perturbation. Macromolecules 40, 6025–6027. 27. Luo, S. C., Xie, H., Chen, N. Y. and Yu, H. H. (2009) Trinity DNA detection platform by ultrasmooth and functionalized PEDOT biointerfaces. ACS Appl. Mater. Interfaces 1, 1414–1419. 28. Lima, A., Schottland, P., Sadki, S., and Chevrot, C. (1998) Electropolymerization of 3,4-ethylenedioxythiophene and 3,4-ethylenedioxythiophene methanol in the presence of dodecylbenzenesulfonate. Synth. Met. 93, 33–41. 29. Gao, Z. Q., Binyamin, G., Kim, H. H., Barton, S. C., Zhang, Y. C., and Heller, A., (2002) Electrodeposition of redox polymers and coelectrodeposition of enzymes by coordinative crosslinking. Angew. Chem. Int. Ed. Engl. 41, 810–813. 30. Lassalle, N., Mailley, P., Vieil, E., Livache, T., Roget, A., Correia, J. P., and Abrantes, L. M. (2001) Electronically conductive polymer grafted with oligonucleotides as electrosensors of DNA: Preliminary study of real time monitoring by in situ techniques. J. Electroanal. Chem. 509, 48–57.

Chapter 28 Preparation and Dynamic Patterning of Supported Lipid Membranes Mimicking Cell Membranes Stefan Kaufmann, Karthik Kumar, and Erik Reimhult Abstract In this chapter, we describe standardized protocols for the self-assembly of supported lipid bilayers (SLBs) from liposomes with lipid compositions mimicking eukaryote and prokaryote cell membranes. Such SLBs can also contain lipids with polymeric and glycosylated headgroups. Furthermore, we present protocols on how to manipulate the adsorption and desorption of membranes on indium tin oxide (ITO) ­electrodes, which allows for the creation of patterned and in situ regenerated SLB arrays that can be used to study electrochemically mediated membrane processes in a microarray format. Key words: Supported lipid bilayer membranes, Liposomes, PEG-lipids, Indium tin oxide, Bio­ sensor, Self-assembly

1. Introduction Supported lipid bilayers (SLBs) are increasingly being used as interfaces in biosensors, biorecognition arrays, and for investigating cell–surface interactions (1). However, the increasing range of application areas demands reproducible ways of producing SLBs with more varied properties in terms of composition (charge, fluidity, headgroup, and tail functionality) and type of solid support material. While the self-assembly of pure supported phospholipid bilayers on glass substrates from liposomes is well established (2), methods for the assembly of membranes with other specialized compositions or containing polymer-modified headgroups have only recently been demonstrated (3–5). Despite the more intricate process of forming SLBs by liposome self-assembly compared to other methods such as painting and solvent spreading, the

Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_28, © Springer Science+Business Media, LLC 2011

453

454

Kaufmann, Kumar, and Reimhult

greater level of control over lipid composition and the absence of all solvents have made this approach increasingly popular. In our efforts to create more natural membrane mimics, we have shown in several cases that it is important to tune the charge interactions between the liposome membrane and the substrate through the introduction of counter ions, a change of pH, or even by dynamically changing the substrate surface potential (4–6). Careful analysis of SLB assembly from complex compositions has created a range of characteristic signatures for successful assembly for ­commonly applied biosensor techniques, but has also highlighted the need for the application of complementary techniques to deconvolute the responses of the most intricate SLB formation processes (5, 7).

2. Materials 2.1. Buffers

1. Tris-buffered saline (TBS): 10  mM Tris(hydroxymethyl) aminomethane (Tris)–HCl, 150 mM NaCl, pH 7.4. Store at 4°C and filter (0.2 mm) before use. 2. TBS with ethylenediaminetetraacetic acid (EDTA): Add 20 mM EDTA to TBS buffer without adjusting the pH. Store at 4°C and filter (0.2 mm) before use. 3. Acetate buffer: 0.1  M Acetic acid, 0.1  M sodium acetate, 150  mM NaCl, pH 5.0. Store at 4°C and filter (0.2  mm) before use. 4. HEPES-buffered saline (HBS): 10 mM 4-(2-hydroxyethyl)1-piperazineethanesulfonic acid (HEPES)–NaOH, 150 mM sodium chloride, pH 7.4. 5. HBS with Ca2+: 10  mM HEPES–NaOH, 150  mM NaCl, 3 mM CaCl2, pH 7.4.

2.2. Liposome Preparation

1. 1-Palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC), dissolved in CHCl3 to 25 mg/ml. 2. 1,2-Dioleoyl-sn-glycero-3-phosphocholine (DOPC), dissolved in CHCl3 to 25 mg/ml. 3. 1,2-Dioleoyl-sn-glycero-3-[phospho-l-serine] (DOPS), dissolved in CHCl3 to 10 mg/ml. 4. 1-Oleoyl-2-[6-[(7-nitro-2-1,3-benzoxadiazol-4-yl)amino] hexanoyl]-sn-Glycero-3-Phosphocholine (NBDPC), dissolved in CHCl3 to 1 mg/ml. 5. 1,2-Dioleoyl-sn-glycero-3-phosphoethanolamine-N[methoxy(polyethylene glycol)(2 kDa)] (PEG-PE), dissolved in CHCl3 to 10 mg/ml.

Preparation and Dynamic Patterning of Supported Lipid Membranes

455

6. 1,2-Distearoyl-sn-glycero-3-phosphoethanolamine-N[poly(ethylene glycol)(2 kDa)-N¢-carboxyfluorescein] (ammonium salt) (PEG-PE CF), dissolved in CHCl3 to 1 mg/ml. 7. Escherichia coli total lipid extract. 8. Lipopolysaccharides (LPS), purified from natural extracts. 9. Round-bottom flask (25 ml). 10. Liquid nitrogen (N2). 11. Liposofast Extruder instrument (Avestin) equipped with polycarbonate extrusion filters (pore size: 200, 100, 50 nm) (Avestin). 2.3. Molecules for Surface Functionalization (Nonfouling)

1. Poly(l-lysine)-graft-poly(ethylene glycol) [PLL(20  kDa)g[3.6]-PEG(2  kDa)] (PLL-g-PEG) (SuSoS AG) diluted to 20 mg/ml in HBS.

2.4. S  ubstrates

1. Silicon wafers: (100) crystal orientation, 10–20 Ω cm resistivity, 500–550-mm thickness, polished. 2. No. 1 glass coverslips (0.13–0.16-mm thickness) (MenzelGläser, Germany) with a refractive index of 1.52 3. No. 1 glass coverslips (0.13–0.16-mm thickness) (MenzelGläser, Germany) or silicon wafers coated with 50 nm TiO2 by magnetron sputtering. 4. No. 1 glass coverslips (0.13–0.16-mm thickness) (MenzelGläser, Germany) and standard microscope glass slides coated with radio-frequency (RF)-sputtered ITO (in Ar atmosphere, 20-nm thick, In2O3:SnO2 = 90:10). 5. ITO microelectrode array: SU8 background interspersed with indium tin oxide (ITO) array spots (40 × 40 mm in size) (Ayanda Biosystems). The ITO array spots should be individually addressable.

2.5. C  leaning Agents

1. Milli-Q water (18.2  MW  cm resistivity). The total organic carbon (TOC) content should not exceed 4 ppm. 2. 2% (w/v) sodium dodecyl sulfate (SDS) in Milli-Q water. 3. Ethanol (absolute, analytical grade, min 99.9% purity). 4. Toluene (analytical grade, min 99.99% purity). 5. Isopropyl alcohol (analytical grade, min 99.9% purity). 6. Chloroform (analytical grade, min 99.9%). 7. Water bath sonicator (35 kHz, 120/240 W). 8. UV/ozone cleaner (UV wavelengths 185 and 254 nm).

456

Kaufmann, Kumar, and Reimhult

2.6. Flow Cells

1. Flow cell enabling full fluid exchange and (for electrochemically controlled dynamic patterning on microelectrode arrays) a three-point electrode contact (working electrode: substrate; reference electrode: Ag/AgCl; counter electrode: Pt).

3. Methods 3.1. Wafer Substrate Surface Preparation 3.1.1. Preparation and Cleaning of Wafer Substrates

1. The silicon wafers may be diced to the desired size. Caution: Wafer dicing can lead to severe contamination of the wafer pieces, depending on the dicing method employed. 2. The wafer pieces should be cleaned by water bath sonication in toluene for 30  min, followed by sonication in isopropyl alcohol for 30 min.

3.1.2. Surface Cleaning Shortly Before Adsorption of Liposomes

1. Before usage, all substrates are cleaned by rinsing or water bath sonication in ethanol.

3.1.3. Reusing Samples

1. If substrates are to be reused, cleaning by water bath sonication in 2% (w/v) SDS and ethanol is recommended before preparing the substrates for SLB assembly by repeating the cleaning steps as described above in Subheading 3.1.2.

3.2. Liposome Preparation

1. A clean round-bottom flask should be thoroughly rinsed with Milli-Q water and dried under a stream of N2.

2. After rinsing in Milli-Q water and drying with a stream of N2 (or other inert clean gas), the wafer substrates are treated in a UV/ozone cleaner for 30 min.

2. Lipids dissolved in chloroform are mixed in the desired concentrations (see Notes 1–3) and dried under a stream of N2 for 30 min. During the initial evaporation of the chloroform, the flask should be constantly rotated to obtain a smooth, solvent-free, thin lipid film. 3. The lipid film is resuspended in the required buffer solution (see Notes 4 and 5), which results in a white turbid suspension of large multilamellar vesicles. A final lipid concentration of 5 mg/ml is suggested, but this may be adjusted according to the experimental requirements from 0.1 to 10 mg/ml. 4. To reduce the multilamellarity of the liposome solution, repeated freeze–thaw cycles are performed before extrusion as follows: The round-bottom flask is introduced into liquid nitrogen until the solution is completely frozen. Next, a warm water bath is used to thaw the solution again. These preceding two steps are repeated a total of five times.

Preparation and Dynamic Patterning of Supported Lipid Membranes

457

5. The starting (pre-extrusion) lipid solution is sucked into one of the two syringe units of the extruder. The extruder is assembled with two stacked porous polycarbonate membranes with the desired pore size. For the production of liposomes with a diameter smaller than 100 nm, two consecutive extrusions are suggested (see Note 6). The solution is extruded 31 times; note that the final solution should be extracted from the syringe unit into which the pre-extrusion solution was not inserted. The final, extruded solution should appear almost transparent if free from multilamellar liposomes. The size distribution of the liposomes should be characterized before use to verify the success of the preparation (see Note 7). 6. The prepared liposome solution is sealed and stored at 4°C under inert gas in a small, airtight glass vial. The solutions should ideally be used within 1 week of preparation, but can be used after several weeks if prepared correctly and stored undisturbed. 3.3. Supported Lipid Bilayer Formation on Wafer Substrates

1. Liposome solutions are diluted in buffer to obtain a 50–100-mg/ml solution (i.e., a 100- or 50-fold dilution of a 5-mg/ml stock solution) (see Notes 8 and 9). If the dilution is made in a buffer that is different from the liposome storage buffer, the diluted suspensions should be equilibrated for 10 min before injection. 2. The liposome suspension is introduced to the cleaned substrates either by constant flow (a sufficient volume of liquid should be used that will last the entire incubation period) or as a static solution. The necessary exposure time of the substrate to the liposome solution depends mainly on the lipid concentration and the composition (see Notes 10–12). 3. The bulk solution (buffer) conditions can be altered by subsequently exchanging the liquid in the flow cell for a different buffer after SLB formation has been completed. Buffers from a large range of conditions may be chosen for exchange; for example, after formation, SLBs deposited at pH 5 can be exposed to buffers at physiological pH without any change in assembled structure.

3.4. SLB Removal and Wafer Substrate Recycling

1. To remove the SLBs, the substrates are rinsed with a 2% (w/v) SDS solution. Depending on the fragility of the substrate, it can be sonicated for 10 min in a 2% (w/v) SDS solution, or incubated in a 2% (w/v) SDS solution for 10  min without sonication. 2. After rinsing thoroughly with Milli-Q water, an additional cleaning step (10 min) in ethanol should also be performed.

458

Kaufmann, Kumar, and Reimhult

3.5. Assembly and Removal of Liposomes and SLBs on ITO Microelectrodes 3.5.1. Assembly of SLBs on an ITO electrode

A method has also been developed by our laboratory to control in situ liposome adsorption, SLB formation, and lipid desorption (regeneration of the substrate) by means of applied currents when ITO electrodes are used as the substrate. This specific protocol is described in the sections below. 1. Prepare a liposome stock solution with a concentration of 5 mg/ml of liposomes with an 80:20 (mol/mol) DOPC:DOPS mixture of lipids suspended in HBS according to the protocol described in Subheading 3.2. 2. Dilute the liposome stock solution in HBS with Ca2+ to a concentration of 0.5 mg/ml. 3. Add the diluted liposome solution to the ITO substrate in a flow cell and incubate for 30 min. 4. Rinse away excess liposomes with HBS.

3.5.2. Blocking the Assembly of SLBs on an ITO Electrode

1. To prevent the adsorption of liposomes to an ITO electrode, the ITO substrate should be incubated with 20 mg/ml PLLg-PEG in HBS for 20 min. This renders the ITO surface to become nonfouling, and prevents any subsequent adsorption of vesicles to the surface.

3.5.3. Removal of SLBs from an ITO Electrode

1. Connect the flow cell to a three-electrode setup, with ITO as the working electrode. 2. Raise the potential on the surface to +1.8  V and apply for 10 min. 3. Rinse the surface with excess HBS to remove the desorbed SLBs.

3.5.4. Formation and Removal of SLBs on an ITO Microelectrode Array (Fig. 1)

1. Assemble the microelectrode array (described in Subhead­ ing 2.4, item 5) with a flow cell unit. 2. Incubate the whole substrate with a solution of 20  mg/ml PLL-g-PEG in HBS for 20 min. 3. Raise the potential of the microelectrode array spot selected to be activated to +1.8 V for 10 min. This removes any PLLg-PEG adhered to the selected microelectrode, but leaves the rest of the substrate passivated. 4. Subsequently introduce the DOPC:DOPS liposomes in HBS with Ca2+ into the flow cell. The vesicles will rupture to form SLBs only on the nonpassivated microelectrodes in the array. 5. The SLBs can subsequently be removed from individual microelectrode array spots by raising the potential of the corresponding microelectrode array spots to +1.8 V. The steps above can be repeated several times for each microelectrode in the array.

Preparation and Dynamic Patterning of Supported Lipid Membranes

459

Fig. 1. Formation and removal of SLBs on ITO microelectrode array spots. (a) The top left and top right spots are coated with PLL-g-PEG while the bottom two spots are coated with supported lipid bilayers (SLBs) containing DOPC:DOPS:NBDPC (78:20:2, mol/mol/ mol). (b) The application of a positive potential (+1.8 V) to the top right electrode removes the PLL-g-PEG. (c) SLBs are assembled on the exposed spot on the top right electrode. (d) Assembled SLBs on the bottom right electrode are removed by the application of a positive potential (+1.8 V). Reproduced with permission from ref. (6) © 2009 Royal Society of Chemistry.

3.6. Characterization of Supported Lipid Bilayers on ITO-Coated Substrates

Although highly reproducible, it is recommended to always characterize the formation of SLBs with one of the following standard techniques: (1) quartz crystal microbalance with dissipation monitoring (QCM-D) (see Note 13); (2) evanescent optical techniques (see Note 14); and (3) fluorescence recovery after photo­bleaching (FRAP) (see Note 15). This is especially important if the standard protocols are modified in any way to accommodate new lipid compositions, substrate preparations, or buffer compositions. To facilitate the characterization process, a set of signature responses for successful SLB formation are described in Notes 13–15 for the most common characterization techniques.

4. Notes 1. The concentration of polymer-lipids is a crucial factor for SLB formation. Depending on the molecular weights of the polymers, the molar concentrations of the polymer-lipids should be chosen to be below the mushroom-to-brush transition for

460

Kaufmann, Kumar, and Reimhult

standard buffers to allow reasonable formation kinetics in the case of PEG-lipids. For further details about the calculation of PEG-lipid concentrations, see Note 3. When glycolipids, EDTA-adjusted or low pH buffers are used, higher molar fractions of polymer lipids can be employed; see Subheading 3.3. 2. When forming negatively charged vesicles necessary for the formation of SLBs on ITO, a mixture of DOPC:DOPS (80:20–70:30, mol/mol) should be used. 3. To calculate the polymer regime, we propose using the theory suggested by De Gennes (8). In this framework, the system is in the mushroom regime for D > RF and in the brush regime for D < RF, where D is the distance between the PEG-lipids and RF is the Flory radius. The Flory radius is calculated by RF ~ an3/5 and the distance between the PEG-lipids is D = A / m , where A is the average area per lipid and m is the mole fraction of PEG-lipids. It should be further noted that the PEG-lipids in the SLB are most likely unevenly distributed between the two leaflets. There are strong indications that the concentration of PEG-lipids in the surface distal leaflet is higher than in the proximal leaflet for a SLB. 4. TBS buffer is preferred for preparing and fusing SLBs from liposomes with pure zwitterionic lipid compositions. 5. Highly negatively charged lipid compositions and bacterial lipid compositions should not be stored in Ca2+-containing buffers. 6. Depending on the pore diameter, repeated extrusion may be needed. For example, to prepare 50-nm liposomes, it is appropriate to extrude through a 200-nm porous filter first and then subsequently re-extrude the solution with 50-nm porous filters. 7. Conventional dynamic light scattering (DLS) measurements are conducted on every batch of vesicles to ensure reproducibility between batches. The nominal size from the extrusion membrane pore diameter is typically not obtained for pore sizes differing from 100 nm. Smaller pore sizes tend to result in liposomes that are larger than predicted and larger pore sizes tend to give smaller liposome diameters than predicted, as measured by DLS. The obtained diameter for a given lipid composition and membrane pore diameter should, however, be consistently reproducible with a low polydispersity. 8. If EDTA-enriched Tris buffer is used, the 20-mM EDTA is added to the Tris buffer solution just before the liposome solution is to be diluted. 9. The formation of SLBs from net negatively charged liposome compositions (including E. coli total lipid extracts) on ITO and TiO2 surfaces requires 3 mM Ca2+-containing buffer.

Preparation and Dynamic Patterning of Supported Lipid Membranes

461

10. Simple mixtures of, e.g., DOPC or DOPC:DOPS liposomes (as well as E. coli liposomes) will typically form SLBs in less than 30 min at the suggested concentrations. 11. E. coli lipid membranes can only be formed on TiO2-coated substrates with the described preparation protocol. TiO2 substrates are also preferred for other strongly negatively charged liposome compositions. 12. The SLB formation time for polymer-lipids depends strongly on the polymer lipid fraction. For PEG-liposomes in the brush regime, incubation for 24 h is recommended. 13. Quartz crystal microbalance with dissipation monitoring (QCM-D). The QCM-D technique monitors changes in the energy dissipation and resonance frequency of a quartz crystal and allows measuring the adsorption of thin films as well as structural properties. The formation of SLBs is characterized by an adsorption phase (increasing dissipation and decreasing frequency) and a fusion phase (decrease of dissipation and increase of frequency). The final frequency shift for a standard 5  MHz crystal is between 25 and 30  Hz (ca. 450 ng/cm2) for most standard lipid compositions, and the dissipation is close to 0 (below 0.2 × 10−6). For polymer-lipid membranes or undulating membranes, the signature kinetics remain the same, but larger remaining frequency and dissipation shifts are expected. Figure  2 shows the QCM-D measurements obtained for the formation of SLBs containing POPC:PEG(MW 2000)-PE in the ratio 96:4, 94:6, 92:8, 90:10 mol%.

Fig. 2. The formation of SLBs containing POPC:PEG(MW 2000)-PE in the ratio 96:4, 94:6, 92:8, 90:10 mol%. The curves are displayed with an offset in the x-axis for clarity.

462

Kaufmann, Kumar, and Reimhult

Fig. 3. The formation of an Escherichia coli total lipid extract SLB on a TiO2 coated waveguides recorded using an evanescent field optical biosensor technique [viz., optical waveguide lightmode spectroscopy (OWLS)]. The kink in the adsorption curve observed at t = 30 min and the sudden saturation of the adsorption at t = 40 min are characteristic for SLB formation processes as recorded by evanescent optical biosensors. Reproduced with permission from ref. (5) © 2008 American Vacuum Society.

14. Evanescent optical techniques. The most common biosensor techniques, such as surface plasmon resonance (SPR), use optical evanescent fields to record changes in the local refractive index at the sensor surface. However, these are not ideal techniques to confirm the adsorption of lipid bilayers since the conformation change from liposomes to planar lipid bilayers is not measured directly. However, an apparent rapid increase in the refractive index close to the surface is recorded with these techniques, as liposome rupture leads to relocation of lipid material closer to the surface and ordering with respect to the surface normal. For the case of liposome adsorption from a static buffer environment, the mass (or refractive index) kinetics results in the characteristic kink in the adsorption curve at t = 30  min in Fig.  3, followed by a sudden stop in adsorption shortly thereafter. 15. Fluorescence recovery after photobleaching (FRAP). Fluore­ scence recovery after photobleaching is used to determine the mobile fraction of lipids in a SLB and to measure the average diffusion coefficient of those lipids. A light source is used to photobleach the fluorophores in the membrane, and then the recovery of the fluorescence intensity in the bleached area is monitored. If the liposomes adsorb but a SLB is not formed, then no recovery of the fluorescence intensity in the bleached area is observed. If the liposomes fuse to a SLB, the quality of the SLB can be estimated from the percentage recovery of fluorescence intensity that is recorded over time in the

Preparation and Dynamic Patterning of Supported Lipid Membranes

463

Table 1 Example FRAP measurements obtained for POPC:PEG(2000)-PE SLBs POPC:PEG(2000)-PE

Labeled compound

Diffusion Recovered coefficient (mm2/s) fraction (%)

96:4

NBDPC

1.0 ± 0.3

96 ± 3

94:6

NBDPC

1.4 ± 0.2

98 ± 3

92:8

NBDPC

0.5 ± 0.2

89 ± 7

90:10

NBDPC

n.r.

n.r.

96:4

PEG-PE CF

2.3 ± 0.4

98 ± 2

94:6

PEG-PE CF

1.2 ± 0.6

97 ± 3

92:8

PEG-PE CF

n.r.

n.r.

n.r. no recovery

bleached area. Table  1 lists the FRAP results measured for SLBs containing POPC:PEG(MW 2000)-PE in the ratio 96:4, 94:6, 92:8, 90:10 mol%. For this evaluation, the theory published by Jönsson et al. (9) was used. References 1. Castellana, E.T. and Cremer, P.S. (2006) Solid supported lipid bilayers: From biophysical studies to sensor design. Surface Science Reports. 61, 429–444. 2. Richter, R.P., Berat, R., and Brisson, A.R. (2006) Formation of solid-supported lipid bilayers: an integrated view. Langmuir. 22, 3497–3505. 3. Albertorio, F., Diaz, A.J., Yang, T., Chapa, V.A., Kataoka, S., Castellana, E.T., and Cremer, P.S. (2005) Fluid and air-stable lipopolymer membranes for biosensor applications. Langmuir. 21, 7476–7482. 4. Kaufmann, S., Papastavrou, G., Kumar, K., Textor, M., and Reimhult, E. (2009) A detailed investigation of the formation kinetics and layer structure of poly(ethylene glycol) tether supported lipid bilayers. Soft Matter. 5, 2804–2814. 5. Merz, C., Knoll, W., Textor, M., and Reimhult, E. (2008) Formation of supported bacterial

lipid membrane mimics. Biointerphases. 3, FA41–FA50. 6. Kumar, K., Tang, C.S., Rossetti, F.F., Textor, M., Keller, B., Vörös, J., and Reimhult, E. (2009) Formation of supported lipid bilayers on indium tin oxide for dynamically-patterned membrane-functionalized microelectrode arrays. Lab on a Chip. 9, 718–725. 7. Reimhult, E., Zäch, M., Höök, F., and Kasemo, B. (2006) A Multitechnique Study of Liposome Adsorption on Au and Lipid Bilayer Formation on SiO2. Langmuir. 22, 3313–3319. 8. de Gennes, P.G. (1987) Polymers at an interface; a simplified view. Advances in Colloid and Interface Science. 27, 189–209. 9. P. Jönsson, P., Jonsson, M.P., Tegenfeldt, J.O., and Höök, F. (2008) A method improving the accuracy of fluorescence recovery after photobleaching analysis. Biophysical Journal. 95, 5334–5348.

Chapter 29 Enzyme Immobilization on Reactive Polymer Films Ana L. Cordeiro, Tilo Pompe, Katrin Salchert, and Carsten Werner Abstract Immobilized enzymes are currently used in many bioanalytical and biomedical applications. This protocol describes the use of thin films of maleic anhydride copolymers to covalently attach enzymes directly to solid supports at defined concentrations. The concentration and activity of the surface-bound enzymes can be tuned over a wide range by adjusting the concentration of enzyme used for immobilization and the physicochemical properties of the polymer platform, as demonstrated here for the proteolytic enzyme Subtilisin A. The versatile method presented allows for the immobilization of biomolecules containing primary amino groups to a broad variety of solid carriers, ranging from silicon oxide surfaces to standard polystyrene well plates and metallic surfaces. The approach can be used to investigate the effects of immobilized enzymes on cell adhesion, and on the catalysis of specific reactions. Key words: Maleic anhydride copolymers, Enzyme coupling, Thin films, Bioactive surfaces, Covalent immobilization

1. Introduction In general, immobilization can improve enzyme stability and selectivity (1), broadening the range of enzyme-based applications in, for example, biocatalysis, analytical devices, and antifouling. Several strategies may be pursued to use enzymes as antifouling agents (2, 3), including the degradation of molecules anchoring the foulers to the surface, and the catalysis of antifouling compounds. The evaluation of the effects of surface-confined enzymes on the settlement, adhesion, and fate of fouling organisms with different adhesion mechanisms will enable further understanding of the molecular processes involved in biofouling. The enzyme immobilization strategy presented here is based on reactive maleic anhydride copolymer films, and can be adapted Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_29, © Springer Science+Business Media, LLC 2011

465

466

Cordeiro et al.

to surfaces of various materials (e.g., glass, polystyrene, poly (dimethysiloxane), and titanium) and with different shapes (2D and 3D) (4–6). The reactive copolymers can be covalently attached as thin films with thicknesses in the nanometer range to amino-functionalized surfaces. The polymer layer does not change other physical parameters of the underlying support, including its micrometer surface roughness and topography. The physicochemical properties of the polymer layer can in turn be varied by the selection of the copolymer used and preparation conditions (4). The described method provides a versatile and molecularly defined platform to attach biomolecules at defined concentrations. It is robust, easy to implement, and can be used in a variety of applications. The immobilization strategy is based on the high reactivity of the anhydride moieties of the copolymer layers toward primary amines. Therefore, the availability of such groups located on the surface of the biomolecule is a fundamental condition for successful immobilization. As such groups are frequently available, this does not appear to be a major hindrance. A limitation of the strategy described is related with the poor control over which amine groups of the proteins bind to the polymer film. If the protein of interest has a scattered distribution of primary amine groups on its surface, this may lead to various orientations of the biomolecule on the polymer film, which in turn decreases the availability of active sites. In the worst-case scenario, the free amino groups are located near the enzyme catalytic (active) site or near a substrate-binding module, hampering enzyme activity upon immobilization. In spite of these limitations, it was recently demonstrated that the immobilization strategy described effectively enables the covalent immobilization of a wide variety of biomolecules (4, 7–9), including enzymes (10), and that the resulting surfaces can be used in a broad range of applications such as, for example, to control the settlement and adhesion of marine foulers (11), or to control stem cell fate decisions (7).

2. Materials 2.1. Preparation of Copolymer Solutions

1. Poly(octadecene-alt-maleic anhydride) 30,000–50,000) (Polysciences, Inc.).

(POMA)

(MW

2. Poly(ethylene-alt-maleic anhydride) (PEMA) (MW 125,000) (Sigma). 3. Tetrahydrofuran (THF) (Caution: Highly flammable, irritates eyes and respiratory system, accumulates in the body, and may form explosive peroxides).

Enzyme Immobilization on Reactive Polymer Films

467

4. n-Hexane (Caution: Highly flammable, irritates skin, and is harmful by inhalation). 5. Acetone (Caution: Highly flammable, irritates eyes, repeated exposure may cause skin dryness or cracking, and vapors may cause drowsiness and dizziness). 2.2. Thin Film Preparation

1. Glass cover slips (24 × 24 mm2) (see Note 1). 2. Maleic anhydride precoated polystyrene well plates (Cresco Biotech) (see Note 2). 3. Isopropanol (Caution: Highly flammable, irritates eyes, and vapors may cause drowsiness and dizziness). 4. Hydrogen peroxide (35% w/w), not stabilized (Caution: Sensitizes skin, irritates mucous membranes, and causes vision loss) (see Note 3). 5. Ammonium hydroxide solution (29% w/w) (Caution: Toxic when inhaled, and causes skin burns). 6. 3-Aminopropyltriethoxysilane (APTES) (Caution: Corrosive, and harmful when swallowed) (see Note 4). 7. PTFE syringe filters, 0.2-mm pore size. 8. Teflon wafer basket (Entegris, Inc.). 9. Glass Petri dishes (25-cm diameter). 10. Quartz-glass beaker. 11. Spin coater (up to 5,000 rpm).

2.3. Enzyme Immobilization

1. Custom-fabricated immobilization chamber (Fig. 1).

2.4. Quantification of Immobilized Enzyme

1. High-performance liquid chromatography (HPLC) system: Agilent 1100 capillary LC system (Agilent Technologies) equipped with a vacuum degasser, a quaternary pump, an autosampler, a reversed-phase column (Agilent Zorbax SB-C18, internal diameter = 4.6 mm, length = 150 mm, particle size = 3.5 mm), and a fluorescent detector (detection and excitation wavelengths: 455 and 355 nm, respectively).

2. Enzyme solution (e.g., Subtilisin A): Dissolve Subtilisin A at a concentration between 0.5 and 30  mg/ml in phosphate buffer saline (PBS) modified by the addition of CaCl2 (2 mM) and NaCl (100 mM). Adjust the pH of the enzyme solution to 8.6 (via addition of NaOH) (see Note 5). Prepare all enzyme solutions only immediately prior to use.

2. Closable 400-ml vessel that can be evacuated (e.g., dessicator). 3. Appropriate glass containers for storage and processing of enzyme-coated glass coverslips.

468

Cordeiro et al.

Fig.  1. The custom-built enzyme immobilization chamber viewed in a disassembled state (a) and after assembly (b). The apparatus conveniently allows for the immobilization of Subtilisin A to well-defined areas of glass coverslips coated with PEMA and POMA copolymer films. The setup consists of a bottom plate and a top frame made of poly(methyl methacrylate) (PMMA). The bottom plate allows the glass coverslips to be accurately positioned. The top frame is placed on the coverslips to fix and seal them for the enzyme immobilization step. At every coverslip position, the top frame has a circular opening of 16  mm in diameter, allowing the application of solutions to the coverslip surface. A seal between the top frame and the coverslips is achieved by rubber O-rings. Using this apparatus, precisely 2 cm2 of the total glass coverslip surface area can be functionalized with enzymes. The setup also ensures that no undefined enzyme attachment occurs on the backside of the coverslips. An additional top cover plate (made of PMMA) may also be used to close the chamber and to avoid contamination and solvent evaporation during the immobilization process.

4. Chromosulfuric acid (Caution: Toxic by inhalation, causes severe burns, may cause cancer, accumulates in the body, and is harmful to aquatic organisms). 5. Redrying reagent: water:ethanol:triethylamine (2:2:1, v/v/v). 6. Gas-phase hydrolysis: 6  M hydrochloric acid containing 1% phenol (w/v). 7. Dissolution buffer: 50 mM sodium acetate buffer, pH 6.8.

Enzyme Immobilization on Reactive Polymer Films

469

8. o-Phthalaldehyde (OPA) reagent: Dissolve 25.2 mg OPA in 500 ml of methanol, 20 ml of 2-mercaptoethanol and 4.5 ml of 0.2 M potassium borate buffer, pH 10.2. 9. Amino acid standards (Sigma–Aldrich). 10. Eluent A: 50  mM sodium acetate buffer (pH 6.8): methanol:THF (80:19:1, v/v/v). 11. Eluent B: Methanol:50 mM sodium acetate buffer, pH 6.8 (80:20, v/v). 2.5. Activity of Immobilized Enzyme

1. Substrate solution: 0.2  mM N-succinyl-Ala-Ala-Pro-PhepNA in PBS, pH 7.4 obtained by the dilution of a 2-mM stock solution in dimethyl sulfoxide (DMSO) (see Note 6). Prepare the substrate solution only prior use. 2. Microplate absorbance reader (405 nm) (see Note 5). 3. Blocking buffer: 0.1 M citric acid. 4. Microtiter plates (96-well).

3. Methods Since the method relies on the reactivity of the maleic anhydride groups of the polymer films toward the primary amines of the enzyme, in order to obtain reliable and reproducible results, it is important to regenerate the anhydride moiety of the prepared polymer films prior to enzyme immobilization via annealing. The quantification of immobilized enzymes can be performed by amino acid analysis based on HPLC following the total hydrolysis of the surface-bound enzyme (12). Ultimately, it is important to determine whether the immobilized enzyme is active. This can be achieved through an appropriate activity assay developed based on the characteristics of the specific enzyme immobilized. The activity assay described in this protocol enables the determination of the activity of the serine protease Subtilisin A (10) (see Note 7). 3.1. Copolymer Solutions

1. To remove olefin impurities, mix POMA and water at a ratio of 1:10 (w/v) and stir the resulting suspension at room temperature overnight. Remove the water by suction filtration and dry the residual polymer at 50°C for 24 h under vacuum. Agitate the dried POMA with n-hexane at a ratio of 1:10 (w/v) at room temperature for 3 h, filter off, and wash with n-hexane. Dry the purified POMA initially at 50°C for 2 h and subsequently at 120°C for 20 h under vacuum. Store the purified polymer at room temperature.

470

Cordeiro et al.

2. Purify PEMA by rinsing the polymer with n-hexane using the procedure described in step 1 above. Store the purified polymer at room temperature. 3. Prepare a 0.16% (w/w) solution of POMA in THF and filter using a syringe filter. Sealed solutions may be stored at room temperature in the dark for 2 weeks. 4. Prepare a 0.3% (w/w) solution of PEMA in a mixture of acetone:THF (1:2, w/w). Filter using a syringe filter. Sealed solutions may be stored at room temperature in the dark for 2 weeks. 3.2. Thin Film Preparation

1. Immerse a Teflon wafer basket accommodating the coverslip substrates into a beaker containing deionized water, and expose it to ultrasound for 30 min. Rinse the samples in the Teflon wafer basket at least three times using not less than 500 ml of clean deionized water each time (see Note 8). 2. Repeat the ultrasonic cleaning process in ethanol, and then rinse the coverslips three times using deionized water as described above (see Note 8). 3. Transfer the coverslips in the Teflon wafer basket directly into a quartz-glass beaker filled with a mixture of hydrogen peroxide (35% w/w), aqueous ammonium hydroxide solution (29% w/w), and deionized water at a volume ratio of 1:1:5 (see Note 8). Heat the mixture to 70°C and maintain at this temperature for 10  min. Do not heat the solution above 70°C. This work has to be performed in a fume hood. Be careful in handling the hydrogen peroxide and ammonia solutions since these are highly corrosive. 4. Rinse the coverslips at least twice in a minimum volume of 500 ml of deionized water (see Note 8). Dry the coverslips with a nitrogen stream (see Note 9). 5. Immediately transfer the substrates into a 20-mM solution of APTES in 90% (v/v) isopropanol (see Note 8). Cover the beaker and wait for exactly 2  h. Rinse the coverslips thoroughly in isopropanol and dry them again with a nitrogen stream. Finally, anneal the coverslips at 120°C for 1  h (see Note 10). 6. Attach a coverslip to the spin coater and apply the PEMA or POMA solution on top of the glass substrate such that liquid covers the entire surface. Immediately (<3 s) initiate rotation of the spin coater using the following conditions: acceleration = 1,500 rpm/s, speed = 4,000 rpm, duration = 30 s. Collect the coverslips in a glass Petri dish while avoiding contact of the polymer-coated surfaces with the Petri dish, and cure at 120°C for 2 h.

Enzyme Immobilization on Reactive Polymer Films

471

7. Upon cooling, immerse the samples in acetone for 10 min. Rinse three times with clean acetone and immediately dry with a nitrogen stream (see Note 11). Samples can be stored in sealed containers at room temperature in the dark for approximately 3  months. A heat treatment (2  h, 120°C), however, is required immediately prior to using the samples if they are stored for a long period of time (see Note 12). 3.3. Enzyme Coupling

1. Prepare sterile enzyme solutions with concentrations between 0.5 and 30 mg/ml (see Note 5). 2. Place the polymer-coated coverslips into the immobilization chamber and add 500  ml of protein solution to each freshly activated maleic anhydride coated surface (the amount of welldefined exposed coating area per sample is 2 cm2) (see Note 13). Incubate at room temperature overnight (see Note 5). 3. Rinse samples with sterile deionized water ten times (see Note 5). 4. Remove samples for HPLC analysis from the immobilization chambers, and store them in appropriate glass vials at −18°C. 5. Samples for measurement of enzymatic activity can be processed directly in the immobilization chambers or in poly­ styrene wells (see Note 14). 6. Maleic anhydride-coated polystyrene well plates can be coated directly in the same way as above without using an immobilization chamber.

3.4. Quantification of Immobilized Enzyme (see Note 15) 3.4.1. Hydrolysis

1. Treat all glassware needed for the hydrolysis reaction with chromosulfuric acid overnight, rinse extensively with deionized water, and dry at 80°C. Be careful when handling chromo­ sulfuric acid, perform the work in a fume hood, and work in a well-ventilated area using safety goggles, appropriate clothing, and gloves. 2. Place the glass vials containing the samples in a 400-ml glass vessel (e.g., dessicator). Include at least two glass vials without samples to be used for blank measurements. Connect the vessel to a vacuum pump and let the samples dry under reduced pressure at room temperature for at least 1 h. 3. Add 4  ml of 6  M hydrochloric acid (HCl) (containing 1% w/v phenol). Close the vessel; expose the samples to vacuum for 15 s and then rinse with nitrogen for 12 s. Repeat the last two steps two times, then remove the nitrogen and apply vacuum. Keep the under-vacuum hydrolysis vessel at 110°C for 24 h. 4. Remove the HCl and add 30 ml of redrying reagent to each vial. Dry the samples under vacuum and store at −18°C until further analysis.

472

Cordeiro et al.

3.4.2. Derivatization

1. Wash out the hydrolysates from the coated glass coverslips by repeated rinsing with 200 ml of 50 mM sodium acetate buffer, pH 6.8. 2. The derivatization of the hydrolysates and standards can be performed using an automated procedure. Mix 30 ml of OPA reagent with 10 ml of sample solution in the HPLC autosampler. Inject 5  ml of the derivatized sample into the HPLC system for binary gradient separation.

3.4.3. Separation

1. These instructions assume the use of an Agilent 1100 capillary LC system equipped with an autosampler, a Zorbax SB-C18 column, and a fluorescence detector. 2. Keep the flow rate at 0.8 ml/min and the column temperature at 30°C. 3. Establish a linear gradient from 0 to 100% eluent B within 30 min. Keep 100% eluent B constant for 3 min, and finally switch to 100% eluent A within 1 min. 4. Include measurements of the amino acid standards (166, 83, 42, 21 pmol). 5. Analyze the chromatograms (i.e., identify and quantify the measured amino acids) using appropriate software (e.g., Agilent Chemstation Software 08.01).

3.4.4. Numerical Analysis

3.5. Activity of Immobilized Enzyme

To calculate the amount of immobilized protein through the measured amino acid amounts, a short (e.g., MATLAB-based) algorithm is needed to solve the linear equation system Ax = B, where vector A corresponds to the amino acid ratios of the proteins under scrutiny, and vector B to the measured amino acid ratios. Amino acid ratios are calculated by normalizing the portion of the amino acid residues to the molecular mass of the protein. Vectors A and B are of length 15 (see Note 16). The best-fit numerical solution of matrix x in the linear equation system is calculated using a least-squares approach (see Note 17). 1. Add 1  ml of substrate solution to the enzyme-containing surfaces at room temperature. Include at least three blank samples without immobilized enzyme. 2. Monitor the elapsed time starting from the moment of substrate addition. 3. Periodically extract aliquots of 95  ml of supernatant (see Note 18) and transfer them into the cavities of a 96-well plate containing 5 ml of 0.1 M citric acid solution. Measure the absorbance of the extracted aliquots at l = 405 nm. 4. Generate a calibration curve by measuring the absorbance at l = 405  nm of standard solutions of p-nitroaniline (pNa) of varying known concentrations.

Enzyme Immobilization on Reactive Polymer Films

473

Fig.  2. Initial activity of Subtilisin A immobilized to POMA and PEMA copolymer films using enzyme solutions of variable concentrations for immobilization [Es]. Reproduced with permission from (10) © 2009 Wiley-VCH Verlag GmbH & Co. KGaA.

5. Calculate the concentration of released pNa for each surface at each time point using the calibration curve. 6. Plot the concentration of released pNa versus time for each enzyme-containing surface (i.e., progress curves of the enzymatic reaction). 7. Determine the initial slope of each reaction curve, designated as the initial surface activity and expressed as [pNa]/min. An example of the initial activity of Subtilisin A immobilized on PEMA and POMA copolymer films using variable concentrations of enzyme during immobilization is shown in Fig. 2.

4. Notes 1. Glass coverslips or silicon wafers of other dimensions may be used. The appropriate sample dimensions depend on the design of the immobilization chamber. For the quantification of immobilized enzymes by amino acid analysis using HPLC, glass cover slips of 24 × 24 mm2 are recommended since those will provide for a large enough surface area to yield immo­ bilized enzyme amounts detectable by HPLC. 2. As an alternative to self-preparing the maleic anhydride copolymer film surfaces in a laboratory, these may be purchased

474

Cordeiro et al.

ready-to-use on polystyrene well plates. The quantification of immobilized enzyme amounts via amino acid analysis using HPLC, however, is not possible using this setup. Alternative methods such as ELISA may be used instead. 3. Do not use stabilized hydrogen peroxide, since stabilized hydrogen peroxide may contain nonvolatile stabilizers (13). 4. 3-Aminopropyltriethoxysilane must be opened under nitrogen (or argon) to avoid contamination with moisture. 5. The immobilization conditions described here are adequate for the immobilization of the proteolytic enzyme Subtilisin A. For the immobilization of other enzymes, the immobilization parameters (enzyme concentration, buffer, pH, temperature, and time of incubation) depend on the characteristics of the enzyme and therefore need to be determined, based on reported optimal conditions, empirically for each enzyme. 6. The substrate must be selected according to the particular enzyme to be surface-immobilized. The substrate solution is experimentally determined by screening different buffers, pH, and substrate concentrations. 7. The activity of immobilized Subtilisin A can be determined by following the enzymatic conversion of the substrate N-succinylAla-Ala-Pro-Phe-pNA into peptides and p-nitroaniline (pNa) by absorbance spectroscopy. 8. During all cleaning and rinsing steps, make sure that the glass coverslips are always fully covered by the surrounding solution. The liquid volume required depends on the receptacle used. The coverslips should be kept in the wafer basket  all throughout experimental steps 1–5 of Subheading 3.2. 9. It is essential to process the glass substrates immediately after cleaning to avoid recontaminating the glass surface. 10. The polymer coating should be performed directly subsequent to aminosilanization due to the susceptibility of the generated surface amino groups to oxidizing substances. 11. Samples must be quickly dried after their removal from the acetone solution. 12. Repeated heat treatment (120°C, 2 h) is required to activate the anhydride moieties if the coated glass coverslips are either stored for long time periods or autoclaved. 13. The volume of added enzyme solution depends on the area of the exposed coating (i.e., the dimensions of the samples used and the exposed coating area in their immobilization chamber). 14. To evaluate the initial enzymatic activity, samples should be measured immediately after preparation.

Enzyme Immobilization on Reactive Polymer Films

475

15. For the quantification of the amount of immobilized enzyme by amino acid analysis using HPLC, glass coverslips with dimensions of 24 × 24  mm2 and immobilization chambers enabling well-defined coated surface areas of 2 cm2 (or higher) are recommended. These conditions should provide for a large enough area to yield immobilized enzyme amounts detectable by HPLC. The coated area needs to be welldefined to enable the accurate calculation of the amount of immobilized enzyme per unit area. For HPLC quantification, the substrates need to be handled with special care throughout the whole process to avoid contamination that could compromise the quality of the results (since both coated and noncoated surfaces of the substrate will be included in the amino acid quantification procedure). All materials used throughout the whole preparation process of aliquots to be measured by HPLC (i.e., sample preparation, hydrolysis, and derivatization) must be thoroughly cleaned to avoid contamination that may negatively affect measurements and calculations. 16. Only 15 amino acids can be detected by HPLC: aspartic acid, glutamic acid, serine, histidine, glycine, threonine, arginine, alanine, tyrosine, methionine, valine, phenylalanine, isoleucine, leucine, and lysine. 17. To improve the results of the calculations, up to three amino acids can be excluded from the calculation when their measured fractions differ by more than 50% from expected values. 18. The periodicity at which each aliquot should be extracted depends on the amount of active enzyme immobilized on each surface and needs to be determined empirically. The volume of the extracted aliquots depends on the total reaction volume being assayed. Ensure good mixing of the supernatant prior to aliquot extraction. References 1. Hanefeld, U., Gardossi, L., and Magner, E. (2009) Understanding enzyme immobilisation, Chem. Soc. Rev. 38, 453–468. 2. Olsen, S. M., Pedersen, L. T., Laursen, M. H., Kiil, S., and Dam-Johansen, K. (2007) Enzyme-based antifouling coatings: a review, Biofouling 23, 369–383. 3. Kristensen, J. B., Meyer, R. L., Laursen, B. S., Shipovskov, S., Besenbacher, F., and Poulsen, C. H. (2008) Antifouling enzymes and the biochemistry of marine settlement, Biotechnol. Adv. 26, 471–481. 4. Pompe, T., Zschoche, S., Herold, N., Salchert, K., Gouzy, M. F., Sperling, C., and Werner, C.

(2003) Maleic anhydride copolymers – A versatile platform for molecular biosurface engineering, Biomacromolecules 4, 1072–1079. 5. Pompe, T., Werner, C., and Worch, H. (2008) Biofunctionalization of titanium substrates using maleic anhydride copolymer films. In Metallic Biomaterials Interfaces (Breme, J., Kirkpatrick, C.J., and Thull, R., eds.), WileyVCH, Weinheim, Germany, pp. 106–109. 6. Cordeiro, A. L., Zschoche, S., Janke, A., Nitschke, M., and Werner, C. (2009) Functionalization of poly(dimethylsiloxane) surfaces with maleic anhydride copolymer films, Langmuir 25, 1509–1517.

476

Cordeiro et al.

7. Alberti, K., Davey, R. E., Onishi, K., George, S., Salchert, K., Seib, F. P., Bornhäuser, M., Pompe, T., Nagy, A., Werner, C., and Zandstra, P. W. (2008) Functional immobilization of ­signaling proteins enables control of stem cell fate. Nat. Methods 5, 645–650. 8. Pompe, T., Salchert, K., Alberti, K., Zandstra, P., and Werner, C. (2010) Immobilization of growth factors on solid supports for the modulation of stem cell fate. Nature Protoc. 5, 1042–1050. 9. Sperling, C., Salchert, K., Streller, U., and Werner, C. (2004) Covalently immobilized thrombomodulin inhibits coagulation and complement activation of artificial surfaces in vitro, Biomaterials 25, 5101–5113. 10. Tasso, M., Cordeiro, A. L., Salchert, K., and Werner, C. (2009) Covalent immobilization

of  Subtilisin A onto thin films of maleic anhydride copolymers, Macromolecular Bioscience 9, 922–929. 11. Tasso, M., Pettitt, M. E., Cordeiro, A. L., Callow, M. E., Callow, J. A., and Werner, C. (2009) Antifouling potential of Subtilisin A immobilized onto maleic anhydride copolymer thin films, Biofouling 25, 505–516. 12. Salchert, K., Pompe, T., Sperling, C., and Werner, C. (2003) Quantitative analysis of immobilized proteins and protein mixtures by amino acid analysis, J. Chromatogr. A 1005, 113–122. 13. Kern, W., and Puotinen, D. A. (1970) Cleaning solutions based on hydrogen peroxide for use in silicon semiconductor technology, RCA Rev. 31, 187–206.

Chapter 30 Characterization of Protein–Membrane Binding Interactions via a Microplate Assay Employing Whole Liposome Immobilization Matthew D. Smith and Michael D. Best Abstract Protein–cell membrane binding interactions control numerous vital biological processes, many of which can go awry during disease onset. However, the study of these events is complicated by the complexity of the membrane bilayer. These efforts would benefit from a rapid and easily accessible method for characteri­zing protein–membrane recognition events. Herein, we describe a microplate-based method for the detection of protein–membrane binding that employs whole liposome immobilization using a biotin anchor. First, control studies are detailed to test for nonspecific liposome immobilization (fluorescence assay; see Subheading 3.2), and to ensure that liposomes remain intact on the microplate surface (dye leakage assay; see Subheading 3.3). Finally, a protein–membrane binding detection assay is described through the example of protein kinase Ca binding to surface-immobilized whole liposomes (see Subheading 3.4). Key words: Chemiluminescence assay, Liposome immobilization, Phospholipids, Protein– membrane binding

1. Introduction Binding events in which soluble receptors form transient noncovalent binding interactions with the surfaces of cellular membranes regulate numerous key physiological and pathophysiological processes (1, 2). These complexation events control both receptor function and subcellular localization, and thus it is of great importance to understand the details of protein–membrane binding at the molecular level. However, a number of challenges exist in this endeavor due to the complex nature of cell membrane bilayers and their cognate receptors. As such, researchers have typically employed simplified systems that mimic cellular membranes. For example, liposomes are particularly effective model systems and Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_30, © Springer Science+Business Media, LLC 2011

477

478

Smith and Best

have been exploited for a range of analytical techniques, including surface plasmon resonance (SPR), monolayer penetration, centri­ fugation, and calorimetric and fluorescence assays (1, 3). Other systems, such as supported lipid bilayers (SLBs), have also been found to be effective (4, 5). However, membrane-binding studies would significantly benefit from a general assay format that utilizes accessible instrumentation and is amenable to efficient highthroughput screening. For this reason, we previously reported the development of a microplate-based assay using optical readout to detect protein binding to immobilized lipid head groups (6) and whole liposomes (7). Herein, we detail a stepwise procedure for the latter approach. This strategy employs whole liposomes immobilized onto streptavidin-coated microplate surfaces via a biotin anchor, followed by protein-binding analysis using chemiluminescence readout via enzyme-linked immunosorbent assay (ELISA). When implementing liposomes for assay development, precautions must be taken to ensure that these bilayers remain intact during analysis. While these model membrane structures can sometimes become compromised when subjected to different types of surfaces, previous studies have shown that the biotin–avidin interaction is particularly effective for the immobilization of liposomes (8–12). This article includes control experiments designed to validate such studies. Specifically, in Subheading  3.2, a fluorescencebased immobilization assay is performed to verify that liposomes are successfully anchored onto the surface of the microplate, and that immobilization is driven by an incorporated biotin anchor rather than through nonspecific binding. In Subheading  3.3, a fluorescence dye leakage assay is conducted to further show that the liposomes remain intact when deposited onto the microplate surface. Finally, once liposome surface attachment has been verified, we describe an assay to detect the binding of a signal transduction enzyme protein to liposome-decorated surfaces (Fig.  1). In this example protocol, we analyzed the binding of protein kinase Ca (PKCa) to liposomes containing a mixture of phosphatidylcholine (PC), phosphatidylserine (PS), and diacylglycerol (DAG). PKCa, whose activity is activated upon binding to cellular membranes, is a key target of interest in biomedicine as it is often implicated in cancer onset due to its role in regulating the cell life cycle (13–15).

2. Materials 2.1. Liposome Preparation

1. L-a-phosphatidylcholine (PC) from chicken egg containing a mixture of fatty acid chains (Avanti Polar Lipids, Inc., Alabaster, AL). Store at −20°C. 2. HEPES buffer: 20 mM HEPES, pH 7.4.

Characterization of Protein–Membrane Binding Interactions via a Microplate

479

Fig. 1. Schematic illustration of a microplate-based assay for the detection of protein–membrane binding using immobilized whole liposomes.

3. CBF buffer: 50 mM 5(6)-Carboxyfluorescein (CBF), 20 mM HEPES, 150 mM NaCl, pH 7.4. 4. Liposome preparation buffer: The composition of this buffer varies depending upon the specific study being performed (see Subheadings 2.2 and 2.4 below). 5. LiposoFast Basic (LF-1) membrane extruder equipped with 200-nm polycarbonate membranes (Avestin, Ottawa, Canada). 2.2. Fluorescence Liposome Immobilization Assay

1. Liposomes are prepared and studied in 20  mM HEPES ­buffer, pH 7.4. 2. Biotin-functionalized membrane anchor 1 (Fig.  2). This biotinylated lipid anchor is used for liposome immobilization, and is synthesized according to the procedure described in ref. (7). Store at −20°C. 3. Fluorescein-functionalized membrane anchor 2 (Fig. 2). This fluorescein–lipid conjugate is used for detection purposes, and is synthesized according to the procedure described in ref. (7). Light-sensitive; store at −20°C.

480

Smith and Best

4. Black Reacti-Bind™ Streptavidin High Binding Capacity (HBC) Coated 96-well microplates (Pierce Biotechnology, Rockford, IL). 5. Synergy 2 multi-detection microplate reader (BioTek, Winooski, VT). 1. 5(6)-CBF. 2. CBF buffer: 50  mM 5(6)-Carboxyfluorescein, 20  mM HEPES, 150 mM NaCl, pH 7.4. 3. 25-mL Plastic syringe packed with Sephadex® G-50 gel filtration chromatography resin. 4. Wash buffer: 20 mM HEPES, 150 mM NaCl, pH 7.4. 5. Detergent solution: 20% (v/v) Triton X-100 in water. 6. Biotin-functionalized membrane anchor 1 (Fig.  2). This biotinylated lipid anchor is used for liposome immobilization, and is synthesized according to the procedure described in ref. (7). Store at −20°C. 7. Black Reacti-Bind™ Streptavidin HBC Coated 96-well microplates (Pierce Biotechnology).

H H N

O NH H

S

HO

O

O

4

O

NH

COOH

4

2.3. Dye Leakage Assay for Ensuring Intact Immobilized Liposomes

O

O O

NH

NH

N N

N N 1

N O

N O

O O 16

O

2

O O 16

O 16

16

Biotin-functionalized membrane anchor

Fluorescein-functionalized membrane anchor

Fig.  2. Synthetic lipid anchors employed for the liposome studies described in this protocol.

Characterization of Protein–Membrane Binding Interactions via a Microplate

2.4. Assay for Detection of Protein–Membrane Binding using Immobilized Liposomes

481

1. Assay buffer: 20 mM HEPES, 100 mM NaCl, 1 mM CaCl2, pH 7.4. (Calcium is necessary for PKCa binding). 2. L-a-phosphatidylserine (PS) from porcine brain containing a mixture of fatty acid chains (Avanti Polar Lipids). Store at −20°C. 3. Diacylglycerol (DAG) (either synthetic or derived from natural sources). 4. Biotin-functionalized membrane anchor 1 (Fig.  2). This biotinylated lipid anchor is used for liposome immobilization, and is synthesized according to the procedure described in ref. (7). Store at −20°C. 5. White Reacti-Bind™ Streptavidin HBC Coated 96-well microplates (Pierce Biotechnology). 6. Human recombinant PKCa, His-tagged from Spodoptera frugiperda (Calbiochem, San Diego, CA). Store at −78°C. 7. Anti-PKCa polyclonal antibody from rabbit (Calbiochem). Store at −78°C. 8. Horseradish peroxidase (HRP)-conjugated anti-rabbit IgG from goat (Chemicon International, Temecula, CA). Store at −78°C. 9. Supersignal ELISA Femto Maximum Sensitivity Substrate (Pierce Biotechnology). Store at 4°C. 10. SigmaPlot scientific data analysis and graphing software (Systat Software, Inc., San Jose, CA).

3. Methods 3.1. Liposome Preparation

1. The lipid compositions and buffers used to form each liposome sample vary depending on the specific experiment being performed. This information is indicated for each assay in the relevant subsections below (see Subheadings 3.2–3.4). 2. Prepare lipid reagent stock solutions by dissolving each lipid compound into chloroform inside a glass vial. While the concentration of each stock solution may vary, representative solutions used in this work are: PC (65.8 mM), PS (6.2 mM), DAG (6.4 mM), Biotin anchor 1 (4.3 mM), Fluorescein–lipid 2 (4.7 mM). 3. Add aliquots of the stock solutions of each type of lipid that compose the liposome to a small glass vial. The amount of each type of lipid stock solution to be added varies depending upon the specific study being conducted (see specific examples below). However, a final total lipid concentration of

482

Smith and Best

5 mM is typically formed and used to calculate the percentage of lipid in each liposome (see Note 1). 4. Remove the chloroform by subjecting the lipid solution to a constant stream of nitrogen. 5. Dry the resulting lipid film under high vacuum overnight. 6. Add an appropriate volume of liposome preparation buffer (typically 500 mL) to the dried lipids to produce a solution with a total lipid concentration of 5 mM. 7. Agitate the suspension of lipid and buffer at 40°C for 40 min. 8. Subject the suspension to ten freeze-thaw cycles. Sample freezing and thawing are performed with a dry ice/acetone mixture, and in 40°C water, respectively. 9. Prepare final liposomes by extrusion (~19 times) using a LiposoFast Basic fitted with a 200-nm polycarbonate membrane to control the liposome size (see Notes 2–5). 3.2. Fluorescence Liposome Immobilization Assay

1. Take and combine aliquots of the stock solutions of lipids in chloroform described in Subheading  3.1, step 2 in the following amounts to form two solutions: (1) Solution A (biotin sample): PC (37.2 mL), biotin anchor 1 (5.8 mL), and fluorescein–lipid 2 (5.3  mL); and (2) Solution B (control sample): PC (37.6 mL) and fluorescein–lipid 2 (4.7 mL). 2. Perform the procedures described in Subheading 3.1, steps 4 and 5, and then rehydrate each lipid mixture in 500  mL of HEPES buffer (pH 7.4) to yield the following solutions (each containing 5 mM total lipids): (1) Solution A: PC (4.9 mM, 98%), biotin anchor 1 (50  mM, 1%) and fluorescein-lipid 2 (50 mM, 1%); and (2) Solution B: PC (4.95 mM, 99%) and fluorescein-lipid 2 (50 mM, 1%). 3. Complete the liposome preparation procedures described in Subheading 3.1, steps 7–9, and dilute each 5 mM liposome stock solution in HEPES buffer (pH 7.4) to generate 12 samples with liposome concentrations ranging from 0 to 300 mM. (Note that the concentration of biotin anchor is lower as it only represents 1% of the liposome). 4. Using a black streptavidin-coated microplate (see Note 6), incubate two rows of wells to be used for analysis with 200 mL of HEPES buffer (pH 7.4) for 30 min. 5. Remove the buffer, and add 100 mL of each liposome sample solution prepared in step 3 above into separate wells of the microplate. Add the samples containing biotinylated liposomes (Sample A) to one row and the nonbiotinylated control liposome samples (Sample B) to the other row. Allow the samples to incubate for 1 h.

483

Relative Fluorescence (F-F0) @ 528 nm

Characterization of Protein–Membrane Binding Interactions via a Microplate 100,000 Liposomes: 98% PC, 1% 1, 1% 2 Liposomes: 99% PC, 0% 1, 1% 2

80,000 60,000 40,000 20,000 0

0

50

100 150 200 [Liposome] Immobilized (µM)

250

300

Fig. 3. Representative results from a fluorescence assay to evaluate biotin-driven liposome immobilization. Reproduced with permission from (7) © 2009 American Chemical Society.

6. Remove all liposome solutions from the microplate and wash each well three times using 250 mL of HEPES buffer (pH 7.4). 7. Add 200 mL of deionized water to each well for fluorescence analysis. 8. Measure the fluorescence intensity in each well using a micro­ plate reader (excitation l = 460 ± 40  nm; emission l = 528 ±  20 nm), with detection from the top of the microplate. 9. Plot the measured fluorescence intensity data against the liposome sample concentrations. The successful immobilization of liposomes is demonstrated by the generation of a monotonically increasing fluorescence versus concentration curve for Sample A (biotinylated liposomes), and the observance of only background signals for Sample B (nonbiotinylated liposomes). Representative data from a fluorescence assay to evaluate biotin-driven liposome immobilization are shown in Fig. 3. 3.3. Dye Leakage Assay for Ensuring Intact Liposomes

1. Take and combine aliquots of the stock solutions of lipids in chloroform described in Subheading  3.1, step 2 in the following amounts: PC (37.6 mL), biotin–lipid 1 (4.7 mL). 2. Perform the procedures described in Subheading 3.1, steps 4 and 5, and then rehydrate the lipid mixture in 500 mL of CBF buffer to yield a solution containing 5 mM total lipids. The lipid content in the resulting solution is as follows: PC (4.95 mM, 99%) and 1 (50 mM, 1%). 3. Complete the liposome preparation procedures described in Subheading 3.1, steps 7–9 to generate liposomes with CBF dye encapsulated within the aqueous interior.

484

Smith and Best

4. Remove unencapsulated CBF by passing the as-prepared liposome solution through a Sephadex® G-50 column equilibrated with wash buffer (see Note 7). 5. Identify chromatography fractions containing liposomes by monitoring the fluorescence signal (excitation l = 490  nm) generated upon the addition of 20 mL of 20% (v/v) Triton X-100 to sample aliquots, which releases encapsulated CBF. Combine the fractions that result in fluorescence to yield a purified working solution of dye-encapsulated liposomes. Estimate the concentration of liposomes in the purified working solution by taking into consideration the original starting amount of liposomes and the final total volume of the combined fractions. 6. Dilute the resulting liposome stock solution with HEPES wash buffer into 12 samples with final liposome concentrations ranging from 0 to 300 mM. 7. Using a black streptavidin-coated microplate (see Note 6), wash a row of wells to be used for analysis with 200 mL of HEPES buffer (pH 7.4) for 30 min. 8. Remove the buffer, and add 100 mL of each liposome sample solution prepared in step 6 above into separate wells of the microplate. Allow the samples to incubate for 1 h. 9. Remove all liposome solutions from the microplate and wash each well three times using 250 mL of wash buffer (20 mM HEPES, 150 mM NaCl, pH 7.4). 10. Add 200 mL of deionized water to each well for fluorescence analysis. 11. Measure the fluorescence intensity in each well using a microplate reader (excitation l = 460 ± 40  nm; emission l = 528 ± 20  nm), with detection from the top of the microplate. 12. To the solution in each well, add 30 mL of detergent buffer (20% v/v Triton X-100) to destroy the liposomes. 13. Perform a second round of fluorescence measurements using the microplate reader (excitation l = 460 ± 40  nm; emission l = 528 ± 20  nm), again with detection from the top of the microplate. 14. Plot the resulting fluorescence data against the liposome sample concentrations. The successful immobilization of dyeencapsulating liposomes is demonstrated by: (1) the observance of low fluorescence intensity signals during the initial measurements due to the CBF molecules being encapsulated within intact liposomes; and (2) the observance of high fluorescence intensity signals in the second round of measurements due to the release of the CBF molecules into

485

Characterization of Protein–Membrane Binding Interactions via a Microplate

Fluorescence @ 528 nm

140,000 Initial fluorescence Fluorescence following detergent

120,000 100,000 80,000 60,000 40,000 20,000 0

0

50

100

150

200

250

300

[Liposome] Incubated (µM) Fig. 4. Representative results from a dye leakage assay to verify the immobilization of intact liposomes. Liposomes incubated with the microplate surfaces initially showed low fluorescence due to encapsulation of the fluorophore molecules within intact liposomes. Only upon detergent addition and liposome rupture did fluorescence become activated. Reproduced with permission from (7) © 2009 American Chemical Society.

s­ olution upon liposome destruction. Representative data from a fluorescent dye leakage assay to evaluate the immobilization of CBF-encapsulating liposomes are shown in Fig. 4. 3.4. Assay for Detection of Protein–Membrane Binding Using Immobilized Liposomes

1. Take and combine aliquots of the stock solutions of lipids in chloroform described in Subheading 3.1, step 2 in the following amounts to form two solutions: (1) Solution A (biotin sample): PC (29.3  mL), PS (80.6  mL), DAG (7.8  mL), and biotin anchor 1 (5.8 mL); and (2) Solution B (control sample): PC (29.6 mL), PS (80.6 mL), and DAG (7.8 mL). 2. Perform the procedures described in Subheading 3.1, steps 4 and 5, and then rehydrate each lipid mixture in 500  mL of assay buffer to yield the following solutions (each containing 5 mM total lipids): (1) Solution A: PC (77%), PS (20%), DAG (2%), and biotin anchor 1 (1%); and (2) Solution B: PC (78%), PS (20%), and DAG (2%). 3. Complete the liposome preparation procedures described in Subheading  3.1, steps 7–9, and dilute the resulting 5  mM liposome stock solution into 12 samples with final liposome concentrations ranging from 0 to 200 mM. 4. Using a white streptavidin-coated microplate (see Note 6), incubate two rows of wells to be used for analysis with 200 mL of assay buffer for 30 min. 5. Remove the buffer, and add 100 mL of each liposome sample solution prepared in step 3 above into separate wells of the

486

Smith and Best

microplate. Add the samples containing biotinylated liposomes (Sample A) to one row and the nonbiotinylated control liposome samples (Sample B) to the other row. Allow the samples to incubate for 1 h. 6. Remove all liposome solutions from the microplate and wash each well three times using 250 mL of assay buffer. 7. Prepare 5 mL of a solution containing 100 ng/mL PKCa in assay buffer. Add 200 mL of this solution to each well of the microplate, followed by a 1-h incubation. 8. After 1  h, remove the protein solution and wash the wells three times with 250 mL of assay buffer. 9. Prepare 5  mL of a solution containing 100  ng/mL rabbit anti-PKCa antibody (see Note 8) in assay buffer. Add 200 mL of this solution to each well of the microplate, followed by a 1-h incubation (see Note 9). 10. After 1 h, remove the antibody solution and wash the wells three times with 250 mL of assay buffer. 11. Prepare 5 mL of a solution containing 25 ng/mL goat antirabbit IgG antibody in assay buffer. Add 200 mL of this solution to each well of the microplate, followed by a 1-h incubation. 12. After 1 h, remove the antibody solution and wash the wells three times with 250 mL of assay buffer. 13. Supersignal ELISA Femto Maximum Sensitivity Substrate consists of two solutions, one with luminol substrate and the other with peroxide. Combine 1.25 mL of each solution to generate the working ELISA substrate solution. 14. Add 100  mL of ELISA substrate solution to each well (see Note 10) and immediately analyze the microplate using a plate reader configured for chemiluminescence detection (emission l = 425 nm detected using a 460 +/- 40 nm filter). Monitor the chemiluminescence signal from each well at constant intervals for up to 10 min. The resulting readings may be averaged (see Note 11). 15. Plot the resulting chemiluminescence data against the liposome sample concentrations using SigmaPlot graphing software to curve fit the data (see Note 12). The successful detection of protein–membrane binding interactions utilizing immobilized liposomes is demonstrated by the generation of a monotonically increasing chemiluminescence versus concentration curve for Sample A (biotinylated liposomes), and the observance of only background signals for Sample B (nonbiotinylated liposomes). Representative data from a chemiluminescence assay to evaluate the binding of PKCa to immobilized liposomes are shown in Fig. 5.

Characterization of Protein–Membrane Binding Interactions via a Microplate

487

Relative Luminescence (L-L0) @ 460 nm

key 77% PC, 20% PS, 2% DAG, 1% 1 78% PC, 20% PS, 2% DAG, 0% 1

350,000 300,000 250,000 200,000 150,000 100,000 50,000 0 0

50

100

150

200

[Liposome] Immobilized (µM)

Fig.  5. Representative results from a chemiluminescence assay to evaluate protein– membrane binding using biotinylated liposome samples (containing biotin anchor lipid 1) and control liposome samples lacking biotin anchor 1. Reproduced with permission from (7) © 2009 American Chemical Society.

4. Notes 1. Nonbilayer lipids (such as DAG) and unnatural lipids (such as 1 and 2) can often only be incorporated into liposomes at low percentages, typically £5–10%. 2. Liposomes are extruded an odd number of times such that the final liposomes end up in the opposite syringe from which they started. 3. In order to prevent the polycarbonate membrane from ripping due to increased pressure during liposome extrusion, proper assembly is required. If ripping of the membrane is observed, the liposomes can be salvaged and reextruded. Ripping of the membrane can be minimized by wetting the filter prior to performing liposome extrusion. 4. Liposomes must be stored at 4°C, and may be used for 1–2 days before they decompose. 5. Liposomes can be analyzed by dynamic light scattering (DLS) to evaluate their size and heterogeneity. 6. The amount of streptavidin loaded onto the microplates can vary, depending upon the commercial source. Thus, one should take into consideration the amount of surface coverage and how it affects the results of the studies.

488

Smith and Best

7. Size exclusion chromatography is conducted by packing a 25  mL plastic syringe with Sephadex® G-50 resin. For this work, gravity was sufficient to allow separation of the liposomes from unencapsulated dye. Prior to loading the liposome samples, proper packing and equilibration of the gel filtration medium was ensured by washing with 100–200 mL of buffer. Care was taken to prevent the top of the gel bed from becoming dry during chromatographic separation. 8. All protein and antibody samples should be aliquoted and stored at −78°C to avoid decomposition due to multiple freeze-thaw cycles. 9. Antibody assays often use blocking additives, such as Tween 20, to avoid nonspecific binding interactions. However, many of these additives are detergents that can rupture liposomes, and thus must be avoided. 10. Air bubbles formed upon the addition of the chemiluminescence substrate solution to the microplate wells can potentially affect the measured chemiluminescence signal intensities. To avoid this issue, any bubbles formed can be ruptured by quickly poking them with pipette tips. 11. To optimize the chemiluminescence signal intensities, the concentrations of each antibody reagent and the sensitivity of the plate reader can be adjusted. 12. Nonlinear regression of the chemiluminescence data may be performed using the Langmuir binding isotherm model.

Acknowledgments This material is based upon work supported by the National Science Foundation (NSF) under CHE-0954297 and DMR0906752. References 1. Cho, W. H., and Stahelin, R. V. (2005) Membrane-protein interactions in cell signaling and membrane trafficking, Ann. Rev. Biophys. Biomol. Struct. 34, 119–151. 2. Lemmon, M. A. (2008) Membrane recognition by phospholipid-binding domains, Nature Rev. Mol. Cell Biol. 9, 99–111. 3. Cho, W. W., Bittova, L., and Stahelin, R. V. (2001) Membrane binding assays for peripheral proteins, Anal. Biochem. 296, 153–161. 4. Chan, Y. H. M., and Boxer, S. G. (2007) Model membrane systems and their applications, Curr. Opin. Chem. Biol. 11, 581–587.

5. Jelinek, R., and Silbert, L. (2009) Biomimetic approaches for studying membrane processes, Mol. Biosyst. 5, 811–818. 6. Gong, D., Smith, M. D., Manna, D., Bostic, H. E., Cho, W., and Best, M. D. (2009) Microplate-based characterization of proteinphosphoinositide binding interactions using a synthetic biotinylated headgroup analogue, Bioconjugate Chem. 20, 310–316. 7. Losey, E. A., Smith, M. D., Meng, M., and Best, M. D. (2009) Microplate-based analysis of protein-membrane interactions via immobilization of whole liposomes containing a

Characterization of Protein–Membrane Binding Interactions via a Microplate ­ iotinylated anchor, Bioconjugate Chem. 20, b 376–383. 8. Liu, X. Y., Yang, Q., Nakamura, C., and Miyake, J. (2001) Avidin-biotin-immobilized liposome column for chromatographic fluorescence on-line analysis of solute-membrane interactions, J. Chromatogr., B 750, 51–60. 9. Vermette, P., Meagher, L., Gagnon, E., Griesser, H. J., and Doillon, C. J. (2002) Immobilized liposome layers for drug delivery applications: inhibition of angiogenesis, J. Cont. Rel. 80, 179–195. 10. Vermette, P., Griesser, H. J., Kambouris, P., and Meagher, L. (2004) Characterization of surface-immobilized layers of intact liposomes, Biomacromolecules 5, 1496–1502. 11. Davidson, W. S., Ghering, A. B., Beish, L., Tubb, M. R., Hui, D. Y., and Pearson, K. (2006) The biotin-capture lipid affinity assay:

489

a rapid method for determining binding parameters for apiloproteins, J. Lipid Res. 47, 440–449. 12. Kalyankar, N. D., Sharma, M. K., Vaidya, S. V., Calhoun, D., Maldarelli, C., Couzis, A., and Gilchrist, L. (2006) Arraying of intact liposomes into chemically functionalized microwells, Langmuir 22, 5403–5411. 13. Gutcher, I., Webb, P. R., and Anderson, N. G. (2003) The isoform-specific regulation of apoptosis by protein kinase C, Cell. Mol. Life Sci. 60, 1061–1070. 14. Griner, E. M., and Kazanietz, M. G. (2007) Protein kinase C and other diacylglycerol ­effectors in cancer, Nat. Rev. Cancer 7, 281–294. 15. Mackay, H. J., and Twelves, C. J. (2007) Targeting the protein kinase C family: are we there yet? Nat. Rev. Cancer 7, 554–562.

Chapter 31 A Bioconjugated Phospholipid Polymer Biointerface with Nanometer-Scaled Structure for Highly Sensitive Immunoassays Kazuki Nishizawa, Madoka Takai, and Kazuhiko Ishihara Abstract This method relates to the preparation of a phospholipid polymer platform and the immobilization of an antibody as a bioaffinity ligand onto the platform to construct a biointerface for highly sensitive immunoassays. The specific phospholipid polymer used in this work is poly[2-methacryloyloxyethyl phosphorylcholine (MPC)-co-n-butyl methacrylate (BMA)-co-p-nitrophenyloxycarbonyl poly(ethylene glycol) methacrylate (MEONP)] (PMBN). The PMBN surface could immobilize specific antibodies through covalent chemical bonding by the reaction between MEONP units and amino groups in the antibody. In addition, the PMBN surface could prevent nonspecific protein adsorption from an analyte sample without the use of blocking reagents based on the fundamental properties of the MPC units. Furthermore, a nanometer-scaled particle deposition surface is constructed with PMBN by an electrospray deposition method to enhance the sensitivity by increasing the overall surface area of the biointerface. Key words: Phospholipid polymer, Biointerface, ELISA, Electrospray deposition

1. Introduction In order to enhance the sensitivity of enzyme-linked immunosorbent assays (ELISA), maintaining the activity of the capture/ detector antibody and the enzyme label, efficiently capturing the target molecules, and suppressing undesired reactions on the substrate are all important. Moreover, to obtain high sensitivity levels for the assay, nonspecific adsorption of analytes, labeled antibodies, and other proteins to the surface should also be suppressed. For this purpose, protein-based blocking reagents such as bovine serum albumin (BSA) and casein are commonly used in laboratories worldwide. However, protein-based blocking reagents denature easily, and the cross-reaction between detection reagents and Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_31, © Springer Science+Business Media, LLC 2011

491

492

Nishizawa, Takai, and Ishihara

blocking reagents remains one of the main causes of a high background and low signal-to-noise ratio in immunoassays. Based on the concept of a protein-free blocking procedure, several types of artificial blocking reagents have been examined (1, 2). Among these, water-soluble amphiphilic 2-methacryloyloxyethyl phosphorylcholine (MPC) polymers have been reported as excellent blocking reagents for ELISA applications (3–5). MPC polymers exhibit high resistance to protein adsorption and effectively suppress the denaturation of biomolecules (6, 7). We have developed a solid-phase biointerface for ELISA by integrating a water-insoluble MPC polymer, namely, poly[MPC-co-n-butyl methacrylate (BMA)-co-p-nitrophenyloxycarbonyl poly(ethylene glycol) methacrylate (MEONP)] (PMBN) (8, 9) as a platform and employing antibodies as bioaffinity ligands. The chemical structure of PMBN is shown in Fig. 1. The characteristics of the MEONP units are suitable for immobilizing biomolecules under very mild physiological conditions, and the biomolecules on the surface exhibited very high activity even after immobilization (10). The PMBN surface effectively suppressed the nonspecific adsorption of proteins based on the intrinsic properties of the MPC units and maintained the activity of the immobilized antibodies even after long-term storage (11). Figure 2 shows a schematic illustration of the PMBN biointerface surface prepared in this research.

Fig. 1. The chemical structure of PMBN. Reproduced with permission from (11) © 2008 American Chemical Society.

Fig.  2. Schematic illustration of the PMBN biointerface surface. The antibodies are immobilized via an oxyethylene chain in the MEONP units present in PMBN. Reproduced with permission from (11) © 2008 American Chemical Society.

A Bioconjugated Phospholipid Polymer Biointerface with Nanometer-Scaled Structure

493

A nanometer-scaled structure having a large surface area is beneficial for the enhancement of specific signals due to the increase in the total amount of immobilized antibodies. Among currently used micro-/nanofabrication methods, electrospray deposition (ESD) has gained much attention as being one of the most promising approaches (12). The advantage of this method is that different types of nano-/microscaled polymer structures ranging from spheres to fibers can be deposited (13). When an ESD substrate is used for the immobilization of biomolecules, the functionality of the polymers can satisfy different performance requirements as the supporting material. Furthermore, the high porosity of ESD substrates and their extremely high surface-area-to-volume ratios can provide large specific surface areas for immobilization (14). While electrosprayed nanostructured surfaces have a significant application potential due to their extremely high surface-areato-volume ratios, the stability of such surfaces is poor under aqueous media due to its metastable phase (causing swelling, etc.) (15, 16). To stabilize electrosprayed structures, several treatments such as crosslinking between polymer chains and blending with other polymers have been carried out (17–19). For applications in immunoassays, an electrosprayed substrate should be stabilized to leverage the advantages offered by the nanostructured surface. With the above considerations in mind, we have developed a stable PMBN-based nanometer-scaled surface prepared by the ESD method as a biointerface surface for immunoassays. In this chapter, protocols for the preparation of PMBN-modified surfaces by a dipcoating method and the ESD method are described.

2. Materials 2.1. Preparation of PMBN-Coated Microtiter Plates

1. PMBN was prepared using MPC, BMA, and MEONP. The MPC and MEONP were synthesized according to previously described methods (3, 8, 9) (see Note 1). BMA was purchased from Nacalai Tesque, Inc. (Kyoto, Japan) and used without further purification. 2. Polymer coating solution: 0.2% (w/w) PMBN in ethanol. The PMBN was dissolved in ethanol at room temperature with gentle stirring. 3. 96-Well polystyrene (PS) microtiter plates (Nunc Maxisorp F8). 4. X-ray photoelectron spectrometer (AXIS-His, Shimadu/Kratos, Kyoto, Japan). The take-off angle of the photoelectrons is 90°.

2.2. Preparation of Nanostructured PMBN Surfaces by Electrospray Deposition

1. Substrate: Glass slides (Matsunami Glass Ind., Ltd., Osaka, Japan). An oxygen plasma treatment using a plasma reactor (PR 500, Yamato Science, Tokyo, Japan) should be carried out before use.

494

Nishizawa, Takai, and Ishihara

2. Polymer solution for ESD: Dissolve PMBN and 1, 4-butylenediamine in ethanol. The final total concentration of PMBN polymer mixed with 1,4-butylenediamine crosslinker in the ethanol solution was adjusted to 5.0% (w/w). 3. Bottomless microtiter plates with separate microwells (300-mL volume capacity) (ProPlate™, Grace Bio Labs, Inc., Bend, OR). 4. Sputter deposition system (SCOTT-C3, Ulvac, Kanagawa, Japan) equipped with a gold (Au) target. 5. Scanning electron microscope (SEM). 2.3. Immunoassay with PMBN Substrates

1. Primary antibody solution: 10 mg/mL mouse antihuman thyroid stimulating hormone (hTSH) IgG (Bioclone Australia Pty Ltd., Sydney, Australia) in a phosphate buffer solution (pH 8.0). 2. Antigen solution: hTSH (Biogenesis Ltd., England, UK) in phosphate-buffered saline (PBS), pH 7.1. 3. Washing buffer: PBS solution containing 0.1% (w/w) Tween 20. 4. Biotin-labeled IgG solution: 0,030 mg/mL biotinylated antihTSH IgG (Bioclone) in PBS containing 1.0% (w/v) BSA) 5. Streptavidin-labeled enzyme solution: 0.16 mg/mL streptavidin–horseradish peroxidase (HRP) (Zymed Laboratories Inc., San Francisco, CA) in PBS containing 1.0% (w/v) BSA. 6. Chromogenic substrate solution: Tetramethylbenzidine (TMB) solution (SUMILON peroxidase chromogenic substrate T, Sumitomo Bakelite Co., Ltd., Tokyo, Japan) 7. Stop solution: 1.0 M H2SO4 solution. 8. Absorbance microplate reader.

3. Methods 3.1. Preparation of PMBN-Coated Microtiter Plates

Coating is performed using a solvent evaporation method (dipcoating) as follows: 1. 300 mL of 0.20% (w/w) PMBN in ethanol is pipetted into a microtiter plate for 10 min. 2. The PMBN solution is removed by inverting the microtiter plate. 3. The microplate is incubated inside a sealed chamber in the same solvent vapor atmosphere at room temperature overnight, and then dried under vacuum overnight. Surface analysis of the PMBN-coated surface should be carried out using X-ray photoelectron spectroscopy (XPS). The XPS

A Bioconjugated Phospholipid Polymer Biointerface with Nanometer-Scaled Structure

495

Fig. 3. X-ray photoelectron spectroscopy (XPS) plots of a PMBN262-n2 dip-coated PS microtiter plate surface. Reproduced with permission from (11) © 2008 American Chemical Society.

spectra of carbon (C1s), oxygen (O1s), phosphorus (P2p), and nitrogen (N1s) are shown in Fig. 3. The C1s spectrum is attributed to the C–N (286.3 eV), C–O (286.7 eV), and C=O (288.8 eV) components from the MPC unit. The spectrum is also attributed to the aromatic carbons based on the p-nitrophenyl ester group (284.5 and 285.8 eV for binding to the nitro group, and 286.3 eV for the ester group). A phosphorus peak and a nitrogen peak attributed to the phosphorylcholine group are also observed. These spectra indicate that the surface of the PMBN-coated PS microtiter plate is covered with the phosphorylcholine groups of the MPC unit and with the p-nitrophenyl ester groups of the MEONP unit. 3.2. Preparation of Nanostructured PMBN Surfaces by Electrospray Deposition

This section describes the preparation of PMBN surfaces by the ESD method. The ESD method can easily increase the surface area of a polymer substrate by the impression of an electric field to create nano-/microscaled spheres and fibers. An example of a typical setup for an ESD device is shown in Fig. 4. A strong electric field (typically up to ~30 kV) is applied between a polymer solution in a syringe with a metal capillary tip and a conductive substrate. When the applied voltage reaches a critical value, the electrostatic forces overcome the surface tension of the drop of polymer solution at the tip of the syringe. The charged droplets become ejected in a jet, which thins under electrohydrodynamic forces, and spray toward the conductive substrate (20, 21). Ultimately, a nano-/microscaled structure is fabricated on the conductive substrate. The morphology of the electrosprayed

496

Nishizawa, Takai, and Ishihara

Fig. 4. Schematic illustration of a typical setup for performing electrospray deposition (ESD) of a polymer onto a substrate.

polymer structure can be controlled by adjusting various parameters, such as applied voltage, feeding rate, the type of polymer, polymer molecular weight, surface tension, conductivity, and viscosity. For application to nanostructured biosensing platforms, the phospholipid polymer, PMBN, is used as an example of a bioconjugated electrosprayed polymer material. 1. To create a conductive substrate for ESD, gold (Au) is sputtered onto a glass slide using a sputtering device. 2. 1,4-Butylenediamine, a diamine compound that can react with the MEONP unit in PMBN, is used as a crosslinker and is added to the PMBN ethanol solution. The ratio of the functional groups present in the reaction mixture, i.e., number of amino groups in 1,4-butylenediamine/number of MEONP units, is 0.50. The final total concentration of PMBN plus 1,4-butylenediamine in the mixed ethanol solution is adjusted to 5.0% (w/w). The solution is kept at room temperature for 2 h to allow the crosslinking reaction to proceed. 3. The rate of the crosslinking reaction can be determined by measuring the absorbance of the PMBN/1,4-butylenediamine mixture in ethanol (see Note 2). 4. The mixed solution is sprayed onto the Au-coated glass slide substrate using an ESD device (esprayer ES-1000, Fuence, Tokyo, Japan) while maintaining a voltage of 20 kV between the polymer solution and the Au surface. 5. The sprayed substrate was heated at 60°C for 10 h to allow for stabilization of the nanostructured polymer surface.

A Bioconjugated Phospholipid Polymer Biointerface with Nanometer-Scaled Structure

497

6. The morphologies of the resulting ESD-modified surfaces can be observed using a SEM. An example of the effects of crosslinking with 1,4-butylenediamine and heat treatment on the stability of electrosprayed PMBN nanostructures in water is shown in Fig. 5.

Fig. 5.  Scanning electron microscope images of PMBN surfaces prepared by ESD (0) without diamine and nonheated, (h0) without diamine and heated, (5) with diamine (amine group/active ester group = 0.5) and nonheated, (h5) with diamine (amine group/active ester group = 0.5) and heat-treated. Each “–w” indicates that the surface was contacted with water for 10 h. A porous structure formed by the deposition of polymer nanoparticles is initially observed on substrates sprayed by ESD (0). However, the structure prepared from polymer solutions without amine compounds collapses its morphology drastically upon immersion in water (0–w). This collapse results from the hydrophilic nature of PMBN, whereby the polymer chains become hydrated and the polymer layer swells. The stability of the electrosprayed PMBN surface in an aqueous medium is improved by crosslinking, but the nanoscale-structural properties of the surface changes when it is immersed in water. Finally, when the PMBN chains are crosslinked with 1,4-butylenediamine compound and then heat-treated, the nanoscale structure of the polymer surface remains fully intact, even when it is immersed in water. Reproduced with permission from (22) © 2010 Elsevier B.V.

498

Nishizawa, Takai, and Ishihara

3.3. Immunoassay with PMBN Substrates

To generate standard curves and quantify the amounts of antigen detected, a sandwich ELISA protocol using the PMBN surfaces prepared by the dip-coating method and the ESD method is described. In this example procedure, hTSH is used as a model antigen. 1. A bottomless microtiter plate containing separate microwells (each having a volume of 300 mL) is attached to a Au-sputtered glass slide whose surface has been coated by PMBN using the ESD method. Alternatively, a PMBN dip-coated PS microtiter plate may also be used for this immunoassay procedure. 2. Mouse anti-hTSH IgG in 150 mL buffer solution is pipetted into the wells of the microtiter plate and allowed to react with the active esters of the PMBN for 24 h at 25°C. 3. The wells are washed three times with PBS. 4. To perform the antigen–antibody reaction, 150  mL of 0–10 mIU/mL hTSH in PBS is pipetted into each well. The plate is incubated for 2 h at 25°C. 5. The wells are washed six times with washing buffer. 6. 150  mL of biotinylated anti-hTSH IgG solution is pipetted into the wells. The plate is incubated for 1 h at 25°C. 7. The wells are washed six times with washing buffer. 8. 150  mL of streptavidin–HRP solution is pipetted into the wells. The plate is incubated for 10 min at 25°C. 9. The wells are washed six times with washing buffer.

Fig. 6.  Standard curves of hTSH generated by ELISA measurements. The squares, circles, and triangles indicate the results with PMBN262-n2 coating, no coating, and BSA blocking, respectively (see Note 4). Error bars represent standard deviation (SD) values, n = 3. With regard to comparing the level of nonspecific background ([hTSH] = 0 mIU/mL), the PMBN262-n2-coated and BSA-blocked surfaces exhibit a lower absorbance than the uncoated surface. Without a blocking treatment, the PMBN262-n2-coated surface reduces the nonspecific adsorption of the labeled IgG to levels similar to that achieved by BSA blocking. Specific signals corresponding to hTSH are observed with PMBN262-n2 coating, and a linear standard curve is obtained. Reproduced with permission from (11) © 2008 American Chemical Society.

A Bioconjugated Phospholipid Polymer Biointerface with Nanometer-Scaled Structure

499

Fig. 7. Absorbance values obtained from the ELISA procedure with electrosprayed PMBN and PMBN/1,4-butylenediamine (amino group/active ester group = 0.5) surfaces (with/without heat treatment) and dip-coated PMBN/BSA-blocked surfaces. The PMBN used for the ELISA measurements is PMBN262-n2. Bars represent SD, n = 3. (*) p < 0.05, (**) p < 0.01. The surfaces without crosslinking and heat-treatment exhibited a small specific signal as compared to the background signal level. This surface can immobilize a large amount of antibodies; however, the immobilized antibodies cannot enhance the specific signal. This suggests that the immobilized antibodies on the surface are buried under the polymer chains because the nanometer-scaled structure of the PMBN surface is destroyed during swelling in the aqueous medium. Both the diamine-crosslinked substrate and the heated substrate, which have a high level of stability in the aqueous medium, exhibited a large specific signal. The sensitivity levels of both the crosslinked and the heattreated PMBN substrates were higher compared to that of the planar PMBN surface generated by dip-coating. Reproduced with permission from (22) © 2010 Elsevier B.V.

10. 100 mL of TMB chromogenic substrate solution is pipetted into the wells and incubated for 20 min at 25°C. 11. 100 mL of stop solution is added into the wells. 12. The optical absorbance of the wells at 450 nm is measured using an absorbance microplate reader. Standard curves for hTSH obtained by ELISA using PMBN dip-coated surfaces are shown in Fig. 6. The specific signal and nonspecific signal depend on the MEONP and MPC unit composition (see Note 3). The ELISA results for a PMBN surface prepared by the ESD method is shown in Fig. 7. Both crosslinking and heat treatment are shown to affect the sensitivity of the ELISA procedure.

500

Nishizawa, Takai, and Ishihara

4. Notes 1. PMBN is synthesized by conventional radical polymerization of MPC, BMA, and MEONP by using 2,2¢-azobisisobutyronitrile (AIBN) as an initiator. After the polymerization reaction, the reaction mixture is precipitated using a mixture of chloroform and diethyl ether (2:8, v/v) as a solvent. The precipitated polymer is collected by filtration with a glass filter and washed with the same solvent used for precipitation, and then dried in vacuum. The ratio of the monomer unit composition in the synthesized PMBN product was determined by 1H-NMR. The molecular weight of PMBN is evaluated by gel filtration chromatography. The results of the synthesis procedure are shown in Table 1. The solubility of the polymer in water depends on the MPC unit composition. PMBN containing more than 40 unit mol% of MPC can be dissolved in water; however, PMBN containing less than 20 unit mol% of MPC is water-insoluble. 2. When the active ester group in MEONP reacts with an amino group and forms an amide group, p-nitrophenoxy ion is released into the solution as a leaving group. The progress of the conversion reaction is followed by the concentration of the released p-nitrophenoxy ion, which absorbs at 310 nm in ethanol and 405  nm in water. In our hands, the reaction achieved equilibrium within 2  h after 1,4-butylenediamine was added to the PMBN ethanol solution. The percentage of the reacted active ester groups in the heated substrate (amine group/active ester group = 0.5) is 65%; however, the residual 35% of the active ester group that remain unreacted was sufficient to permit immobilization of antibodies to the PMBN platform. 3. The results of ELISA assays performed using PS microplate wells dip-coated with PMBN of various compositions (Table  1) demonstrate that the MPC unit composition can affect the level of background signals. An increase in the proportion of MPC units decreased the background levels. This suggests that the MPC units reduced the nonspecific adsorption of secondary labeled IgG. On the other hand, an increase in the proportion of the MEONP units enhanced the level of specific signals. Due to an increase in the number of available binding sites for primary antibodies, the overall amount of immobilized antibodies increased, leading to an enhancement of the specific signal. 4. To prepare the BSA-blocked and uncoated surfaces, primary antibody solution was pipetted into the wells of a naked

A Bioconjugated Phospholipid Polymer Biointerface with Nanometer-Scaled Structure

501

Table 1 Characterization of PMBN synthesized by conventional radical polymerization of MPC, BMA, and MEONP by using 2,2¢-azobisisobutyronitrile (AIBN) as an initiator Monomer unit composition (mol%) In feed Sample no.

In compositionb

Mwc Yield Solubilityd

n a MPC BMA MEONP MPC BMA MEONP (104) (%)

In water In ethanol

PMBN631-n2 2

60

30

10

55

42

3

4.8

77

+

+

PMBN451-n2 2

40

50

10

38

59

3

4.3

80

+

+

PMBN271-n2 2

20

70

10

18

78

4

3.0

72

−

+

PMBN262-n2 2

20

60

20

22

65

13

3.0

58

−

PMBNI72-n2 2

10

70

20

16

74

10

1.6

34

−

+

PMBN262-n4 4

20

60

20

19

72

9

3.4

72

−

+

PMBN262-n8 8

20

60

20

–

–

2.7

57

−

+

–

[monomer] = 1.0 mol/L, [AIBN] = 10 mmol/L in ethanol; temperature, 60°C; reaction time, 6 h a  Number of repeating units in the oxyethylene chain of MEONP b  Determined by 1H-NMR spectrum c  Determined by GPC in water/ethanol = 3/7, PEO standards d  Solubility was determined with 10 mg/mL of each polymer sample and described as soluble (+) and insoluble (−) at 25°C

microtiter plate. After incubation for 24 h at 4°C, the wells were washed three times with PBS. Next, 300 mL of 1.0% (w/v) BSA in PBS solution for BSA-blocked surface and PBS solution for uncoated surface were added into separate wells. After incubation for 24 h at 4°C, the wells were again washed three times with PBS.

Acknowledgments A research Fellowship for Young Scientists from the Japan Society for the Promotion of Science (2010632) supported one of the authors (KN) for carrying out this research. This work was also supported by a Grant-in-Aid for Scientific Research on Innovative Areas “Molecular Soft-Interface Science” from the Ministry of Education, Culture, Sports, Science and Technology of Japan.

502

Nishizawa, Takai, and Ishihara

References 1. Engvall, E., Perlmann, P. (1972) EnzymeLinked Immunosorbent Assay, Elisa III. Quantitation of Specific Antibodies by Enzyme-Labeled Anti-Immunoglobulin in Antigen-Coated Tubes. J. Immunol. 109, 129–135. 2. Nagasaki, Y., Kobayashi, H., Katsuyama, Y., Jomura, T., Sakura, T. (2007) Enhanced immunoresponse of antibody/mixed-PEG coimmobilized surface construction of high-performance immunomagnetic ELISA system. J. Colloid Interface Sci. 309, 524–530. 3. Ishihara, K., Ueda, T., Nakabayashi, N. (1990) Preparation of Phospholipid Polymers and Their Properties as Polymer Hydrogel Membranes. Polym. J. 22, 355–360. 4. Sakaki, S., Iwasaki, Y., Nakabayashi, N., Ishihara, K. (2000) Water-Soluble 2-Methacryloyloxyethyl Phosphorylcholine Copolymer as a Novel Synthetic Blocking Reagent in Immunoassay System. Polym. J. 32, 637–641. 5. Ishihara, K., Takai, M. (2009) Bioinspired interface for nanobiodevices based on phospholipid polymer chemistry. J. R. Soc. Interface 3, S279-S291. 6. Sibarani, J., Takai, M., Ishihara, K. (2007) Surface modification on microfluidic devices with 2-methacryloyloxyethyl phosphorylcholine polymers for reducing unfavorable protein adsorption. Colloid Surf. B: Biointerfaces 54, 88–93. 7. Ishihara, K., Nomura, H., Mihara, T., Kurita, K., Iwasaki, Y., Nakabayashi, N. (1998) Why do phospholipid polymers reduce protein adsorption? J. Biomed. Mater. Res. 39, 323–330. 8. Park, J.-W., Kurosawa, S., Watanabe, J., Ishihara, K. (2004) Evaluation of 2-Methacryloyloxyethyl Phosphorylcholine Polymeric Nanoparticle for Immunoassay of C-Reactive Protein Detection. Anal. Chem. 76, 2649–2655. 9. Konno, T., Watanabe, J., Ishihara, K. (2004) Conjugation of Enzymes on Polymer Nanoparticles Covered with Phosphorylcholine Groups. Biomacromolecules 5, 342–347. 10. Goto, Y., Matsuno, R., Konno, T., Takai, M., Ishihara, K. (2008) Polymer Nanoparticles Covered with Phosphorylcholine Groups and Immobilized with Antibody for High-Affinity Separation of Proteins. Biomacromolecules 9, 828–833.

11. Nishizawa, K., Konno, T., Takai, M., Ishihara, K. (2008) Bioconjugated phospholipid polymer biointerface for enzyme-linked immunosorbent assay. Biomacromolecules 9, 403–407. 12. Burger, C., Hsiao, B. S., Chu, B. (2006) Nanofibrous materials and their applications. Annu. Rev. Mater. Res. 36, 333–368. 13. Renker, D. H., Chun, I. (1996) Nanometer diameter fibers of polymer, produced by electrospinning. Nanotechnology 7, 216–223. 14. Kim, J., Grate, J. W., Wang, P. (2006) Nanostructures for enzyme stabilization. Chem. Eng. Sci. 61, 1017–1026. 15. Yao, L., Haas, T. W., Guiseppi-Elie, A., Bowlin, G. L., Simpson, D. G., Wnek, G. E. (2003) Electrospinning and Stabilization of Fully Hydrolyzed Poly(Vinyl Alcohol) Fibers. Chem. Mater. 15, 1860–1864. 16. Liu, Y., Cui, L., Guan, F., Gao, Y., Hedin, N. E., Zhu, L., Fong, H. (2007) Crystalline Morphology and Polymorphic Phase Transitions in Electrospun Nylon 6 Nanofibers. Macromolecules 40, 6283–6290. 17. Zhang, Y. Z., Venugopal, J., Huang, Z.-M., Lim, C. T., Ramakrishna, S. (2006) Crosslinking of the electrospun gelatin nanofibers. Polymer 47, 2911–2917. 18. Lee, S. J., Oh, S. H., Liu, J., Soker, S., Atala, A., Yoo, J. J. (2008) The use of thermal treatments to enhance the mechanical properties of electrospun poly(−caprolactone) scaffolds. Biomaterials 29, 1422–1430. 19. Sangsanoh, P., Supaphol, P. (2006) Stability Improvement of Electrospun Chitosan Nanofibrous Membranes in Neutral or Weak Basic Aqueous Solutions. Biomacromolecules 7, 2710–2714. 20. Hohman, M. M., Shin, M., Rutledge, G., Brenner, M. P. (2001) Electrospinning and electrically forced jets. II. Applications. Phys. Fluids 13, 2221–2236. 21. Yarin, A. L., Koombhongse, S., Reneker, D. H. (2001) Taylor cone and jetting from liquid droplets in electrospinning of nanofibers. J. Appl. Phys. 90, 4836–4846. 22. Nishizawa, K., Takai, M., Ishihara, K. (2010) Stabilization of phospholipid polymer surface with three-dimensional nanometer-scaled structure for highly sensitive immunoassay. Colloid Surf. B: Biointerfaces 77, 263–269.

Part V Biofunctionalization of Nanostructures

Chapter 32 Purification, Functionalization, and Bioconjugation of Carbon Nanotubes John H.T. Luong, Keith B. Male, Khaled A. Mahmoud, and Fwu-Shan Sheu Abstract Bioconjugation of carbon nanotubes (CNTs) with biomolecules promises exciting applications such as biosensing, nanobiocomposite formulation, design of drug vector systems, and probing protein interactions. Pristine CNTs, however, are virtually water-insoluble and difficult to evenly disperse in a liquid matrix. Therefore, it is necessary to attach molecules or functional groups to their sidewalls to enable bioconjugation. Both noncovalent and covalent procedures can be used to conjugate CNTs with a target biomolecule for a specific bioapplication. This chapter presents a few selected protocols that can be performed at any wet chemistry laboratory to purify and biofunctionalize CNTs. The preparation of CNTs modified with metallic nanoparticles, especially gold, is also described since biomolecules can bind and self-organize on the surfaces of such metal-decorated CNTs. Key words: Carbon nanotubes (CNTs), Gold nanoparticles, Purification, Noncovalent/covalent biofunctionalization

1. Introduction Since their discovery (1), carbon nanotubes (CNTs) have attracted enormous attention owing to their unique structural, mechanical, and electronic properties (2–5). CNTs are also one of the few molecules that are known precisely at the atomic level with respect to the physical location of each atom. Arc-vaporization (6, 7) of two pure graphite rods (~1 mm apart) under helium, argon, or – even better – with methane (8) at low pressure (50–700 mbar) produces multiwalled carbon nanotubes (MWCNTs) with diameters of 2–100 nm. This mesoscale graphite system typically consists of two to ten incommensurate concentric cylinders of graphitic shells

Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_32, © Springer Science+Business Media, LLC 2011

505

506

Luong et al.

with a layer spacing of 0.3–0.4 nm. When the anode graphite rod contains a metal catalyst (Fe or Co), single-walled carbon nanotubes (SWCNTs) are generated instead of MWCNTs (9). SWCNTs are truly a single large molecule comprised of a cylindrical graphite sheet of nanoscale diameter capped by hemispherical ends with a typical diameter of 1  nm. Double-walled CNTs (DWCNTs) have also been synthesized and commercialized recently. The lengths of both SWCNTs and MWCNTs can range from hundreds of nanometers to upwards of 20 cm. CNTs thus display extremely high aspect ratios (length/diameter); and, indeed, the aspect ratio often exceeds 10,000 in most preparations. Together with their pure hexagonal structure, CNTs are composed entirely of an sp2 bond structure, which is even stronger than the sp3 bond structure in diamond. Consequently, CNTs are virtually insoluble in most solvent systems. Nevertheless, surfactants such as sodium dodecyl sulfate are able to coat CNTs with negative sulfonate groups that are exposed to the surrounding environment. And Triton X-100, a nonionic detergent, also coats CNTs with their hydrophilic groups oriented toward the aqueous phase. Consequently, CNTs can be dispersed in an aqueous medium in the presence of ionic or nonionic surfactants. The drawback of the arc-discharge method is the costly removal of non-nanotube carbon and metal catalyst materials. An alternative synthesis method, laser ablation, can be adapted for fullerene and SWCNT production by focusing a CO2 laser beam on a target with a high boiling temperature, e.g., carbon composites doped with catalytic metals. The target is vaporized in a hightemperature argon atmosphere, and SWCNTs formed are conveyed via a gas stream to a collection chamber. The temperature, metal species, and the gas flow rate can all affect the resulting diameter of the nanotubes (10). Furthermore, large-scale production is a critical issue with this method, despite the high homogeneity of the SWCNTs produced. The chemical vapor deposition (CVD) synthesis method uses a hydrocarbon vapor, which is thermally decomposed at the surface of a metal catalyst to form the nanotubes. In this method, CNT growth is governed by the type of hydrocarbon gas and/or catalyst used and the growth temperature. Unlike the arc-discharge and laser ablation methods, the CVD approach allows for control over the location and the alignment of the synthesized CNTs (11). Owing to the symmetry and unique electronic structure of graphene, the structure of a CNT strongly affects its electrical properties. SWCNTs are one-dimensional conductors, wherein all of the electrons are confined to move within a single atomic layer and all of the atoms in the SWCNT structure are surface atoms. To date, MWCNTs have received less attention than SWCNTs since their structures are more complex (each carbon

Purification, Functionalization, and Bioconjugation of Carbon Nanotubes

507

shell of a MWCNT can exhibit different electronic properties and chirality). Nevertheless, MWCNTs combine a very similar morphology and set of properties as compared to SWCNTs, while improving significantly their resistance to chemicals. This feature is especially important when the functionalization of CNTs is required for bioconjugation and bioapplications as discussed later. In the case of SWCNTs, covalent functionalization will break some C=C bonds, leaving “holes” in their structure that can affect both their mechanical and optoelectronic properties. In contrast, only the outermost wall of MWCNTs is modified during such functionalization reactions. Remarkable advances have been made in the synthesis and functionalization of CNTs over the past decade, and this has helped to arouse considerable interest in potential applications of CNTs in the fields of biomedical materials, drug delivery, and tissue engineering. In addition, the quest for probing important analytes with very low detection limits and high specificity using biosensors incorporating nanoscale components has intensified, with several avenues being explored. In particular, CNTs have become an extremely popular theme in recent bioapplication research because of their nanoscale diameter, high electrocatalytic activity, and decreased vulnerability to surface fouling. CNTs can also be utilized as novel electrode materials; owing to their high surfaceto-volume ratio, the variation of their electronic conductance to adsorbed surface species could potentially provide a sufficient level of sensitivity for single-molecule detection. Moreover, on the basis of their well-defined structure and high surface area, CNTs appear to be ideal materials for studying interactions with biomolecules, such as proteins, receptors, enzymes, etc. CNTbased sensors are also more stable than metal oxide sensors since they are not affected by mass transfer phenomena or by chemical changes to the surface carbon atoms. Thus, CNTs are promising materials for electrochemical sensors and show great potential for use in the next generation of biosensor devices. And finally, functionalized CNTs may also find useful applications in medicinal chemistry. The use of CNTs as drug delivery scaffolds and substrates for vaccines is feasible, and CNTs functionalized with bioactive moieties are well suited for targeted drug delivery applications, since they exhibit a high propensity to traverse cell membranes with low cytotoxicity. Like graphite, CNTs are relatively inert except at the nanotube caps, which are more reactive due to the presence of dangling bonds. To date, several strategies have been devised to solubilize, purify, and functionalize CNTs. For bioconjugation applications, the most successful approach is to functionalize sp2 carbons at the sidewalls of nanotubes with organic pendant groups. Another important procedure is the noncovalent functionalization of

508

Luong et al.

CNTs through supramolecular interactions (e.g., p–p stacking interactions), which allows the formation of stable suspensions of the nanostructures. This chapter presents some selected protocols to purify and bioconjugate CNTs that can be performed at any wet chemistry laboratory.

2. Materials 2.1. Carbon Nanotubes

Both pristine and chemically modified CNTs can be obtained from various commercial sources with high purity (see Note 1). The user should request a copy of the “Product Specifications” from the supplier, which states the purity, dimensions, and the size-length distribution of the CNTs. If possible, transmission electron microscopy (TEM) images and the Raman signatures of the CNTs should also be obtained from the supplier.

2.2. General Chemicals and Solvents

1. Concentrated sulfuric acid: Although nearly 100% (w/w) sulfuric acid can be prepared, this loses SO3 at the boiling point to produce a solution that is only 98.3% (w/w) sulfuric acid. The ~98% (w/w) grade is more stable in storage, and is the usual form of what is typically described as “concentrated sulfuric acid” (approximately 18 M). 2. Concentrated nitric acid: Nitric acid is miscible with water, and distillation gives an azeotrope with a concentration of 68% (w/w) HNO3 and a boiling temperature of 120.5°C at 1  atm. Two solid hydrates are known; the monohydrate (HNO3·H2O) and the trihydrate (HNO3·3H2O). Nitric acid is isoelectronic with the bicarbonate ion. The concentrated nitric acid of commerce consists of the maximum boiling azeotrope of nitric acid and water. Technical grades are normally 68% (w/w) HNO3 (approximately 15  M), while reagent grades are specified at 70% (w/w) HNO3. 3. Concentrated hydrochloric acid: Commercial aqueous HCl is 35–38% (w/w) or approximately 11.5–12.4  M, respectively. 4. N-N-Dimethylformamide (DMF). 5. Dimethyl sulfoxide (DMSO). 6. Toluene. 7. Sodium dodecyl sulfate (SDS). 8. Ethanol. 9. Chloroform. 10. Methanol. 11. Anhydrous tetrahydrofuran (THF).

Purification, Functionalization, and Bioconjugation of Carbon Nanotubes

509

12. Acetone. 13. Acetic acid. 14. Sodium chloride. 15. Triton X-100. 16. Glycerol. 17. Ethylene glycol. 18. 2-Mercaptoethanol. 19. Calcium carbonate. 20. Sodium tetraborate. 21. 11-Aminoundecanoic acid. 22. Ammonium iron sulfate. 23. Potassium hydroxide. 24. Ammonium peroxodisulfate. 25. Sodium hydroxide. 26. Hydrogen peroxide. 27. Sodium borohydride. 28. Cystamine (2,2¢-diaminodiethyl disulfide). 29. Thionyl chloride. 2.3. Buffers and Other Specialty Chemicals for Coupling and Bioconjugation

1. 1-Ethyl-3-(3-dimethylaminopropyl)-carbodiimide (EDC). 2. 1-Pyrenebutanoic acid, succinimidyl ester (PASE). 3. Poly(diallyldimethylammonium chloride) (PDAC, 20% w/w in water, MW 100,000–200,000). 4. Poly(sodium 4-styrenesulfonate) (PSS, MW 70,000). 5. Polyethyleneimine (PEI, Mn = 423). 6. 1-(3-Aminopropyl)-3-methylimidazolium bromide (IL-NH2). 7. Nafion perfluorinated ion-exchange resin, 5% (w/w) aqueous dispersion. 8. Glutaraldehyde (25% w/w in water). 9. Gold(III) chloride trihydrate (HAuCl4·3H2O). 10. Positively charged gold nanoparticles. 11. a-Cyclodextrin. 12. N-Hydroxysulfosuccinimide (Sulfo-NHS). 13. Pyrene aldehyde. 14. N-(3-(Trimethoxysilyl)propyl)ethylenediamine (AEAPTMS, (CH3O)3SiCH2CH2CH2NHCH2CH2NH2). 15. 3-Aminopropyltriethoxysilane (APTES). 16. Ethylenediamine.

510

Luong et al.

17. Dithiobis(N-succinimidyl propionate) (DTSP). 18. 50  mM (N-morpholino)ethanesulfonic acid (MES) buffer solution, pH 5.7. 19. Phosphate-buffered saline (PBS) solutions, pH 6.8–7.5. 20. 88 mM sodium periodate solution in deionized water. Protect from light. 21. Biomolecules for conjugation: Various oxidases/reductases useful for the construction of biosensing platforms are commercially available. Other biomacromolecules of interest such as proteins, receptors, and peptides can also be obtained from various commercial sources. 2.4. Other Materials and Small Equipment

1. Zirconium oxide (100 nm particle size). 2. Track-etched polycarbonate membrane filter (0.4  mm pore size). 3. Filter paper (0.45 mm pore size). 4. Fritted glass support for filtration (47 mm diameter). 5. Filter membrane (100 nm pore size). 6. Nylon membrane (0.22 mm pore size). 7. Ultrafiltration membrane (50 kDa MWCO). 8. Polycarbonate membrane filter (0.22 mm pore size). 9. Dialysis membrane (10 kDa MWCO). 10. TEM grids. 11. Silicon wafers. 12. Indium tin oxide (ITO) wafers. 13. Quartz wafers. 14. Glassy carbon electrodes (3 mm in diameter). 15. Ultrasonic probe sonicator. 16. Ultrasonic water bath sonicator. 17. Centrifuge. 18. Centrifuge tubes. 19. Hotplates and stirring motors. 20. Laboratory oven for baking at 120°C 21. Vacuum oven for drying. 22. Reflux apparatus. 23. Ice bath. 24. UV–visible spectrophotometer. 25. Magnet (0.1 T).

Purification, Functionalization, and Bioconjugation of Carbon Nanotubes

511

3. Methods 3.1. Purification of CNTs

As-prepared CNTs contain carbonaceous impurities, such as amorphous carbon and graphite nanoparticles, and particles of the ­transition-metal catalysts (see Note 2). The use of high-temperature oxidation in air is effective in removing amorphous and graphitic contaminants from MWCNTs; for SWCNTs, however, metal catalysts must first be removed before this oxidation step since such metals are known to catalyze the low-temperature oxidation of CNTs (12, 13). In principle, the purification of SWCNTs is feasible by using a combination of the following: gas- or vaporphase oxidation; wet-chemical oxidation/treatment; and centrifugation or filtration (including chromatography techniques) (13–24). These procedures, however, require several steps using special setups and instrumentation at elevated ­temperatures (800–1,200°C). If CNTs with high purity are not available, the following two protocols described below can be used to purify SWCNTs and MWCNTs.

3.1.1. Purification of SWCNTs

SWCNTs are often produced using metal catalysts such as cobalt, nickel, iron, or their mixtures; therefore, the final products will typically contain such metallic nanoparticles (10–20 nm) embedded within a capsule of several graphene sheets. The magnetic properties of SWCNTs are thus overwhelmed by the presence of these ferromagnetic materials; and since the magnetic particles strongly adhere to the bundles and are not easily dispersed, there is great importance in purifying as-prepared SWCNTs. The usual treatment of the SWCNT soot with strong acids cannot eliminate these metallic clusters without attacking the nanotubes themselves. Furthermore, purification methods using high temperature are also not desirable since they tend to collapse the nanotubes, thus changing their dimensions. On the other hand, SWCNTs can be purified mechanically using inorganic nanoparticles in an ultrasonic bath, which removes the particles from their graphene shells. Permanent magnetic poles are then used to trap the released particles (25). 1. Prepare a concentrated suspension of SWCNTs in either a SDS detergent (soap) solution or toluene. 2. Disperse the suspension in a solvent such as DMF (see Note 3) or 30% (w/w) nitric acid. 3. Add a powder of inorganic nanoparticles such as zirconium oxide (100 nm in size) or calcium carbonate to the suspension. 4. Sonicate the slurry for 24 h using a water bath sonicator. Trap the magnetic particles released during the sonication procedure

512

Luong et al.

onto the side of a container using a permanent magnet of 0.1 T. Retain the remaining SWCNT slurry. 5. Dissolve and remove the zirconium oxide or calcium carbonate from the slurry by additional acid treatment. 3.1.2. Purification of MWCNTs

MWCNTs produced by the arc-discharge method typically have a wide distribution of lengths and diameters. They also tend to be heavily contaminated with nanoparticles (up to 50% by weight) consisting of nested, closed graphitic layers of a polyhedral shape. Purification of these nanotubes by oxidation in either the gas or liquid phase frequently leads to damage at the openings of the tube ends as well as at the sidewalls. Furthermore, solvents such as methanol, ethanol, or acetone are unable to provide a stable suspension, as aggregation (~100  mm in size) quickly occurs. However, this aggregation can be overcome by the use of surfactants such as SDS, leading to a simple protocol for the separation of the MWCNTs from such contaminated nanoparticles (26). 1. Prepare a suspension containing 2.5 g of SDS and 50 mg of raw MWCNTs in 500  mL of distilled water. Sonicate the suspension for 15 min. (For this protocol, the SDS concentration was optimized to be twofold higher than the critical micelle concentration, CMC = 0.0082  M in pure water at 25°C). 2. Sediment and centrifuge the suspension at 5,000 × g for 10 min to remove large (>500 nm) graphitic particles. Recover the supernatant containing a colloidal suspension of the MWCNTs. 3. Filter the colloidal suspension (containing nanotubes) through a track-etched polycarbonate membrane (0.4-mm pore size) placed on a 47-mm diameter fritted glass support. During the filtration process, sonicate the suspension with an ultrasonic probe placed 5 mm above the filter (see Note 4). Collect the filtrate (containing primarily nanoparticle contaminants) and recover the residue (containing primarily CNTs) from the filter by scraping. 4. Resuspend the recovered CNT residue in an SDS solution (as in Step 1) and repeat Steps 2 and 3 to further improve the purity of the MWCNT sample (see Note 5). Alternatively, the MWCNTs can also be purified without using filtration steps as follows: 1. Prepare a suspension containing 15 g of SDS and 50 mg of raw MWCNTs in 500 mL of distilled water. Sonicate the suspension for 15 min. (For this protocol, the starting SDS concentration was optimized to be 12-fold higher than the CMC.)

Purification, Functionalization, and Bioconjugation of Carbon Nanotubes

513

2. Sediment the suspension for a few hours and collect the aggregates (containing CNTs). The suspension contains smaller nanoparticle contaminants. 3. Resuspend the sediment in a solution containing 10 g of SDS in 500 mL of distilled water (SDS concentration = 8× CMC). Sediment the suspension for a few hours and collect the aggregates. 4. Resuspend the sediment in 7.5 g of SDS in 500 mL of distilled water (SDS concentration = 6× CMC). Sediment the suspension for a few hours and collect the aggregates for further purification if necessary. 3.2. Noncovalent Sidewall Functionalization of SWCNTs for Protein Immobilization (see Notes 6 and 7)

3.2.1. Protein Immobilization to Pyrene-Functionalized SWCNTs Deposited on a TEM Grid or Silicon Wafer

CNTs must first be derivatized either covalently or noncovalently followed by bioconjugation to achieve the best stability, accessibility, and selectivity of the conjugated biomolecule of interest. The noncovalent functionalization of CNTs allows their sp2 bond structure – and thus their important electronic characteristics – to be preserved. In this approach, the heterobifunctional molecule, PASE is irreversibly absorbed to SWCNTs, as the pyrenyl group interacts strongly with the hydrophobic surface of the SWCNTs due to p-stacking. The succinimidyl ester group located at the other end of the PASE molecule is highly reactive toward nucleophilic substitution with the free amines typically found on the surfaces of most proteins, resulting in the formation of an amide bond (Fig. 1). This concept has been successfully used to immobilize biomolecules to SWCNTs on gold grids (27) (see Subheading 3.2.1). Similarly, glucose oxidase can be immobilized via noncovalent functionalization of SWCNTs deposited on a silicon wafer to form a sensitive pH sensor (28). The protocol can also be used to attach SWCNTs to oxide substrates, such as quartz, SiO2 on Si, and ITO (29, 30) (Subheading 3.2.2). However, in this scheme the succinimidyl ester moiety of PASE is first used to link the molecule to an amino-modified substrate; this is then followed by the noncovalent attachment of SWCNTs to the available pyrenyl groups. And finally, as discussed in Subheading 3.2.3, noncovalent functionalization can also be used to attach magnetic nanoparticles to SWCNTs through a carboxylic derivative of pyrene rather than the succinimide derivative (31). 1. Incubate a TEM grid (27) or silicon wafer (28) with a sample of SWCNTs in a solution containing 6 mM PASE in DMF for 1–2 h at room temperature. (NB: Alternatively, the solution can be prepared at 1 mM in methanol (27).) 2. Wash the SWCNT-modified grid or wafer three times with pure DMF (or methanol) to wash away excess PASE reagent. 3. Add an aqueous protein solution to the functionalized SWCNTs for 18  h at room temperature. For example, add

514

Luong et al. O N

O

O

O O

N O

O O

Non-covalent binding

H2N H N O

Biomolecule

Covalent binding

Fig. 1. Noncovalent functionalization of single-walled carbon nanotubes (SWCNTs). The bifunctional molecule, 1-pyrene­ butanoic acid, succinimidyl ester is irreversibly absorbed to SWCNTs since the pyrenyl group interacts strongly with the hydrophobic surface of the SWCNTs via p-stacking. The succinimidyl ester groups are highly reactive to nucleophilic substitution with the amines found on the surfaces of most proteins, resulting in the formation of an amide bond.

ferritin at 5  mg/mL in a 7.5-mM NaCl solution (27) or 10 mg/mL glucose oxidase in filtered deionized water (28). 4. Rinse the TEM grid or wafer thoroughly with pure water for 6 h to remove excess protein. 5. Dry the grid or wafer and store at 4°C (see Note 8). 3.2.2. Noncovalent Immobilization of SWCNTs to Pyrene-Functionalized Oxide Substrates

1. Clean the oxide substrate, such as quartz, SiO2 on Si, or ITO, with a 1:1 (v/v) mixture of concentrated HCl/methanol for 30 min, and then rinse with water. 2. Immerse the cleaned oxide substrate in concentrated sulfuric acid for 30 min to maximize the number of surface hydroxyl groups, and then rinse with water. 3. Boil the acid-cleaned oxide substrate in water for several minutes, and then blow dry with a stream of N2 gas. 4. React the surface hydroxyls of the clean oxide substrate with a 1% (v/v) solution of a heterobifunctional linker molecule

Purification, Functionalization, and Bioconjugation of Carbon Nanotubes

515

(i.e., an organosilane containing a free amino group on one end and a silane group at the opposite end, such as, e.g., AEAPTMS) in 1 mM acetic acid for 20 min at room temperature, resulting in a surface covered with amino functional groups. Rinse three times with water, blow dry, and then bake the substrate for 3–4 min at 120°C. 5. Mix together 500 mL of 0.1 M sodium tetraborate (pH 8.5) with 400 mL of DMF containing 1 mg (2.9 mM) of PASE. Immerse the substrate containing free amino groups into this solution for 12 h, resulting in a pyrene-covered surface. 6. React the pyrene-functionalized surface for 2 h with ­oxidized SWCNTs (prepared as described in Subheading  3.3) (see Note 9). 3.2.3. Immobilization of Capped Magnetic Nanoparticles to SWCNTs Functionalized with a Carboxylic Acid Pyrene Derivative (31)

1. Prepare the carboxylic acid pyrene derivative (PyAH) by dispersing 1 mmol of pyrene aldehyde and 1 mmol of 11-aminoundecanoic acid in 30  mL of absolute ethanol. Stir the reaction mixture at room temperature for 24 h and separate the solution from unreacted pyrene aldehyde by filtration. Evaporate the ethanol slowly and precipitate the characteristic yellow Schiff base. 2. Disperse 1 mg of purified SWCNTs (refluxed in 2.5 M HNO3 for 12 h followed by washing with water and drying) in a solution of 5  mg of carboxylic acid pyrene derivative (obtained from Step 1) in 50 mL of chloroform/ethanol (1:1, v/v). Stir overnight until the yellow color of the supernatant disappears. 3. Prepare capped magnetic iron oxide nanoparticles by adding 1.14 g of KOH in 20 mL of distilled water to 2 g of ammonium iron sulfate in 50 mL of water. Add 0.38 g of ammonium peroxodisulfate in 10 mL of water and 1.5 mL of oleic acid in 30  mL of toluene. Heat the mixture at 80°C for 30  min. Separate the organic phase and precipitate the capped iron oxide nanoparticles with ethanol. 4. Add 5  mg of capped magnetic iron oxide nanoparticles (obtained from Step 3) in 5 mL of chloroform to the PyAHfunctionalized SWCNTs and stir for 2 days (see Note 10). 5. Separate the solution from the residue by filtration and wash with acetone.

3.3. Preparation of Oxidized Carbon Nanotubes for Covalent Functionalization with Biomolecules

In order to covalently attach a biomolecule to CNTs, the introduction of reactive functional groups on the CNTs is a prerequisite. The carboxylic acid group is widely used since it is easily formed on CNTs via oxidizing treatments. The oxidatively introduced carboxyl groups present useful sites for further modifications, as they enable the covalent coupling of molecules through the creation of amide and ester bonds. By this method, a wide range of functional

516

Luong et al.

moieties may be incorporated onto nanotube surfaces, for which purpose bifunctional molecules (e.g., diamines) are often utilized as linkers. A few illustrative examples of this approach include ­nanotubes equipped with dendrimers, nucleic acids, enzymes, metal complexes, or semiconductors and metal nanoparticles. The oxidation of CNTs can be achieved by extensive ultrasonic treatment in a mixture of concentrated nitric and sulfuric acid (Subheading 3.3.1) (32). Such treatment leads to the opening of the nanotube caps with holes formed in the tube sidewalls, followed by oxidative etching along the walls. The resulting CNTs that are obtained typically display lengths ranging from 100 to 300  nm, with both ends and the sidewalls decorated predominantly with a high density of carboxyl groups. In an alternative oxidation protocol (Subheading 3.3.2), nanotube shortening can be minimized by refluxing the CNTs in concentrated nitric acid alone (33). Using this approach, the resulting CNTs have most of their pristine mechanical and electrical properties preserved; however, oxidation only occurs at the opening of the tube caps and carboxylic acid functional groups are only formed at defect sites found along the sidewalls. 3.3.1. Oxidation of Carbon Nanotubes in a Concentrated Sulfuric Acid/Nitric Acid Mixture

1. Suspend 10 mg of CNTs in 40 mL of a 3:1 (v/v) mixture of concentrated H2SO4/HNO3 in a 100-mL test tube and sonicate in a water bath for 24 h at 35–40°C (see Note 11). 2. Dilute the resultant suspension with 200  mL of distilled water, and collect the (larger sized) oxidized CNTs on a filter membrane (100-nm pore size). 3. Wash the collected CNTs extensively with a 10-mM NaOH solution. 4. Polish the cut tubes by suspending them in a 4:1 (v/v) mixture of concentrated H2SO4/30% aqueous H2O2 with stirring at 70°C for 30  min. This step will result in CNTs containing carboxylic acid groups rather than carboxylate groups. 5. After washing in water and filtering on a filter membrane (100-nm pore size), resuspend the cut CNTs at a concentration of 0.1 mg/mL in a 0.5% (w/w) Triton X-100 solution.

3.3.2. Oxidation of Carbon Nanotubes in Concentrated Nitric Acid

1. Reflux 8.5 g of CNTs in 1.2 L of 2.6 M nitric acid for 45 h. 2. Cool the solution, and then transfer it to polytetrafluoroethylene centrifuge tubes and spin at 2,400 × g for 2 h. 3. Decant the supernatant acid and replace it with deionized water. Shake vigorously to resuspend the solids, followed by a second centrifugation–decanting cycle. 4. Resuspend the collected solids in 1.8 L of water with 20 mL of Triton X-100 surfactant, and adjust the mixture to pH 10 with sodium hydroxide.

Purification, Functionalization, and Bioconjugation of Carbon Nanotubes

517

3.4. Preparation of Metallic Nanoparticle/ CNT Composites

In order to deposit CNTs onto electrodes, their hydrophobic nature should be reduced by introducing hydrophilic groups onto the nanotube surface (34). Numerous methods have been reported in the literature for the functionalization and dispersion of CNTs (35). The important factors necessary for a good dispersion include having intact CNT structures without any major damage or modifications, and achieving a uniform homogeneous solution with long-term stability. The dispersion of CNTs is typically carried out by using physical methods (e.g., ultrasonication and milling) and chemical-based approaches (via covalent or noncovalent functionalization). Among the various creative methods that have been reported, we have selected a few representative protocols for the preparation of CNT/nanoparticle composites. The incorporation of nanoparticles into CNT-based materials can lead to an enhanced level of electrocatalytic activity for many electrochemical processes, potentially giving rise to several highly efficient applications such as photoelectrochemical cells and sensor devices. Gold or platinum nanoparticles (NPs), e.g., can be attached to the walls of CNTs by various approaches, including physical evaporation, chemical reaction with functionalized CNTs, and layer-by-layer deposition methods (36–40). This section describes some key protocols for the preparation of thiolated CNTs, and the self-assembly of gold and other metallic nanoparticles onto such thiol-modified CNTs. The gold nanoparticles can be easily prepared with controlled size and dispersion in the presence of a-cyclodextrin, as described in Subheading 3.4.5 (41–43).

3.4.1. Preparation of MWCNT/AuNP Composites by Mechanical Processing and One-Step Citrate Reduction in Aqueous Solution (44, 45)

1. Add potassium hydroxide (200 mg) and MWCNTs (10 mg) to a ruby mortar and grind them together for 1 h at room temperature. 2. Dissolve the reaction mixture in double-distilled, deionized water (10 mL) and precipitate several times in methanol to remove the potassium hydroxide. 3. Sonicate the MWCNTs in 10 mL of water for 6 h to obtain a uniform dispersion. The preceding steps make the MWCNTs hydrophilic in nature and help break down larger bundles of MWCNTs into smaller ones (44). 4. Prepare a suspension of the MWCNTs (1 mg; obtained from Step 3) in a 1% (w/w) HAuCl4 solution (1 mL) by sonication for 5 min to make the nanotubes disperse evenly throughout. 5. Dilute the MWCNT/HAuCl4 suspension to 100  mL with double-distilled water and heat to boiling while stirring. 6. To obtain MWCNTs coated with gold nanoparticles, add 1.5 mL of 1% (w/w) sodium citrate to the boiling MWCNT/ HAuCl4 solution, and continue heating for 5–10  min until the color of the solution does not change (45).

518

Luong et al.

3.4.2. Preparation of MWCNT/AuNP Composites by Electrostatic Layer-byLayer Assembly (39)

1. Heat pristine MWCNTs in air at 600°C for 2 h and then soak in hydrochloric acid for 24 h and centrifuge. 2. Rinse the resulting precipitate with deionized water and dry under air. 3. Chemically functionalize the MWCNTs by ultrasonication in a mixture of sulfuric acid/nitric acid (3:1, v/v) for 8 h. 4. Centrifuge and wash the acid-treated MWCNTs with deionized water three times and oven dry. Redisperse the MWCNTs (containing carboxylic acid groups) in deionized water. 5. For layer-by-layer assembly, prepare a solution of PDAC (MW 100,000–200,000) at a concentration of 0.1  mg/mL in deionized water containing 0.05 M NaCl. Similarly, prepare a solution of 0.1 mg/mL PSS(MW 70,000) in deionized water containing 0.05 M NaCl. 6. Disperse the MWCNTs in the PDAC solution (0.1 mg/mL) for 3  h with sonication to obtain MWCNTs coated with a layer of PDAC. 7. Mix the PDAC-modified MWCNTs with the PSS solution (0.1  mg/mL) for 1  h with sonication, in order to obtain MWCNTs coated with a PDAC/PSS bilayer. 8. Centrifuge and wash the polymer-modified MWCNTs three times with deionized water. 9. Add a solution containing positively charged gold nanoparticles to the dispersion of polyelectrolyte-coated MWCNTs to obtain self-assembled MWCNT/AuNP complexes.

3.4.3. In Situ Synthesis of AuNPs on PSS Polymer-Wrapped MWCNTs Using Ionic Liquids (46)

1. Disperse 10.5 mg of purified MWCNTs with 0.70 g of PSS in 70  mL of distilled water with continuous sonication for 15  min, and then hold the temperature at 50°C for 12  h under vigorous agitation. 2. Remove excess PSS by performing three centrifugation/ resuspension cycles, spinning at 12,000 × g for 30  min each time. Dry the product under vacuum at 60°C overnight to obtain the MWCNT/PSS product as a powder. 3. Dissolve 42 mg of IL-NH2 (72) in 11.6 mL ultrapure water, and then add 0.20 mL of an aqueous solution of MWCNTs/ PSS (2 mg/mL) dropwise into the mixture under stirring to form a well-dispersed solution. 4. To prepare the MWCNT/polymer/gold nanoparticle-ionic liquid (MWCNT/PSS/Au-IL) composites, add 0.20 mL of HAuCl4 aqueous solution (20  mM) dropwise over several minutes to the above mixture (MWCNTs/PSS). 5. After stirring for 10  h, filter the mixture through a nylon membrane filter (0.22  mm pore size). Wash the recovered

Purification, Functionalization, and Bioconjugation of Carbon Nanotubes

519

MWCNT/PSS/Au-IL composite product thoroughly with water, and then dry overnight at 60°C under vacuum. 3.4.4. Deposition of AuNPs on PEI Polymer-Wrapped MWCNTs (47)

1. Disperse 0.4 mg of purified MWCNTs (containing carboxylic acid groups) in 15  mL of distilled water under ultrasonic treatment. 2. Transfer 1.5 mL of the MWCNT dispersion into a vial, and then add 0.04–0.1 mg of cationic PEI polymer using a 1.0 M PEI aqueous stock solution. Mix the solution by a combination of vigorous stirring and sonication. 3. Add 0.2 mL of HAuCl4 (24.3 mM) to the above mixture to give the desired molar ratio of PEI (monomer unit) to gold. Previous work has shown that the packing density of the attached gold NPs generally increases as the initial molar ratio of PEI:HAuCl4 decreases (i.e., from 400:1 to 9.8:1) (47). 4. Heat the resulting mixture at 60°C for 20 min. 5. After washing and centrifugation for two cycles, disperse the resulting MWCNT/AuNP composite product in 1  mL of water.

3.4.5. Preparation of SWCNT/AuNP Composites by Borohydride Reduction in the Presence of Cyclodextrin (41–43)

1. Disperse 1 mg of purified, acid-treated SWCNTs into 2 mL of deionized water and sonicate for 2 h to generate a black suspension of nanotubes. 2. Add HAuCl4 (1  mM, 0.8  mL) and a-cyclodextrin (a-CD) (2 mM, 1 mL) to the SWCNT suspension. 3. Add 20-mL aliquots (to a maximum volume of 100  mL) of freshly prepared, ice-cold 0.1 M sodium borohydride (NaBH4) with slow mixing using a magnetic stir bar. 4. Continue stirring the reaction mixture until a stable reddish color is observed. 5. Incubate the solution for 24 h at room temperature (~22–24°C) to let it stabilize. 6. Purify the AuNP/SWCNTs by washing and centrifugation (8,000 × g) three times to remove the unbound AuNPs. The supernatant should appear clear after the first washing cycle.

3.5. Bioconjugation of Carbon Nanotubes

Bioconjugated CNTs have numerous potential biomedical ­applications based on their unique properties. Some examples of such applications include biosensing, cell tracking and labeling, tissue engineering (as scaffolds), as well serving as vehicles for the intracellular transport of drugs, genes, and/or proteins. In ­particular, the efficient intracellular delivery of anticancer drugs, siRNA, proteins, and DNA combined with their corresponding effective therapeutic effects suggests that CNTs may provide a promising approach for transporting bioactive molecules in the

520

Luong et al.

next generation of biomedicines. CNTs may also serve as ­important components in novel bioengineered nanomaterials that can potentially impart several important features, such as greater mechanical strength per unit mass, good electrical and thermal conductivity, and the ability to guide cell growth and tissue regeneration. CNTs can also be used as a platform for probing ­surface–protein or protein–protein interactions toward the development of highly specific biomolecule detectors. Indeed, it is anticipated that the next wave of advances in the nanobioelectronics field is the synthesis and fabrication of high-density, multiplexed nanotube device microarrays for proteomic applications for simultaneously detecting large numbers of different ­target proteins in “real world” samples. Although the noncovalent modification of CNTs with a pyrene moiety can be used to conjugate a biomolecule of interest as discussed earlier (see Subheading 3.2.1), the specific bioconjugation of target biomolecules is most commonly achieved by covalent methods. Another interesting procedure is the use of APTES to solubilize CNTs to form a very stable CNT/ APTES complex. Biomolecules can then become covalently conjugated to the amino group of APTES via glutaraldehyde activation. This procedure, described below in Subheading 3.5.1, has been used to prepare enzyme electrodes with efficient direct electron transfer (48). 3.5.1. Solubilization and Bioconjugation of MWCNTs using APTES (48)

1. Dissolve 2.0 mg of MWCNTs in a mixture containing 100 mL of APTES, 100 mL Nafion perfluorinated ion-exchange resin and 800 mL of ethanol. 2. Sonicate the mixture for 20  min to obtain a suspension of uniformly dispersed MWCNTs. 3. Drop 3 mL of the MWCNT solution onto the surface of a solid hydrophobic substrate, e.g., a glassy carbon (GC) electrode (3 mm in diameter) and dry in air to form a uniform film containing a network of MWCNTs. 4. Drop 3 mL of a solution of enzyme (e.g., 20 mg/mL glucose oxidase in 50  mM phosphate buffer, pH 7.2) onto the MWCNT/APTES-modified GC electrode and dry in air. (Other enzymes can easily replace the glucose oxidase used here to form biosensors for other analytes.) 5. Drop 3 mL of a 2.5% glutaraldehyde solution (diluted tenfold from a 25% (w/w) stock solution in 50 mM phosphate buffer, pH 7.2) onto the electrode to crosslink the enzyme with APTES and dry. 6. Store the enzyme-modified electrode at 4°C in 50 mM phosphate buffer, pH 7.2.

Purification, Functionalization, and Bioconjugation of Carbon Nanotubes 3.5.2. Covalent Coupling of Carboxylated CNTs with Biomolecules Using EDC/ NHS (see Note 12 and Fig. 2)

521

1. Prepare a 5-mL solution of 5 mM EDC and 10 mM sulfo-NHS in 50 mM MES buffer solution (pH 5.7) containing 0.5 M NaCl (see Notes 13 and 14). 2. Suspend carboxylated CNTs (1 mg/mL) in 5 mL of 50 mM MES buffer solution containing 0.5 M NaCl (pH 5.7). 3. Add the solutions from Steps 1 and 2 together (to give a total volume of 10  mL) and let the resulting mixture react for 30 min at room temperature. 4. Remove the excess reactants from the EDC/sulfo-NHS-activated CNTs by successive centrifugation and dilution with deionized water, followed by filtering and concentrating with an ultrafiltration membrane (50 kDa MWCO). 5. Dissolve the protein to be coupled at a concentration of 1–10 mg/mL in 500 mL of 0.1 M phosphate buffer, 0.5 M NaCl, pH 7.2. 6. Resuspend the activated CNTs from Step 4 in the buffered protein solution prepared in Step 5. 7. Allow the mixture to react under continuous agitation with a magnetic stirrer for at least 2 h at room temperature. 8. Add 2-mercaptoethanol to the reaction solution to obtain a final concentration of 20 mM to quench the coupling reaction. Alternatively, if the protein being activated is sensitive to this concentration level of 2-mercaptoethanol, quench the reaction by performing the following: wash the mixture by extensive centrifugation/resuspension using a suitable buffer (e.g., 10  mM sodium phosphate, 0.15  M NaCl, pH 7.4),

Fig. 2.  Carbon nanotubes containing carboxylate groups (COOH-CNTs) can be coupled to amine-containing biomolecules using an aqueous two-step coupling process based on EDC/NHS (or sulfo-NHS) chemistry to form an amide bond.

522

Luong et al.

followed by dialysis using an ultrafiltration membrane (50 kDa MWCO). 9. Resuspend the conjugated CNTs in 10 mM sodium phosphate, 0.15 M NaCl, pH 7.4, and store at −20°C. In this protocol, the carboxyl groups of CNTs are to be converted to amino groups by refluxing the CNTs in DMF in the presence of thionyl chloride. The resulting CNTs are modified with ethylenediamine to form an intermediate that can react with aldehyde groups, which can be introduced into the biomolecules of interest through various methods. For example, as described below, enzymes and proteins that are glycosylated (e.g., glucose oxidase and horseradish peroxidase) can be oxidized with sodium periodate to produce aldehyde groups for reductive amination coupling to amino-modified CNTs (Fig. 3).

3.5.3. Conversion of CNT Carboxyl Groups to Amino Groups for Covalent Bioconjugation (49)

1. Place oxidized MWCNTs (containing carboxyl groups after treatment with sulfuric and nitric acids; see Subheading 3.3.1) in a vacuum oven and dry at 70°C overnight. 2. Suspend the dried CNTs (1  mg/mL) in anhydrous DMF using an ultrasonic water bath. 3. Add the CNT dispersion to thionyl chloride and reflux for 24 h to convert the carboxylic acids to acyl chlorides. 4. Wash the resulting nanotubes with anhydrous THF (five times) to remove excess SOCl2 using a polycarbonate membrane filter (0.22 mm pore size). 5. React the nanotubes with ethylenediamine for 3 days under continuous stirring to form CNTs with amino groups. 6. Wash the amino-modified CNTs with deionized water extensively to remove excess ethylenediamine using a polycarbonate membrane filter (0.22 mm pore size). 7. Prepare a solution of a carbohydrate group-containing enzyme of interest in PBS buffer (e.g., glucose oxidase in PBS, pH 6.8).

O

O

C

C OH

SOCl2/DMF Reflux

O Cl SH-(CH2)2-NH2

C N

SH

H

Pyridine/DMF

Fig.  3.  Conversion of carboxylate group-containing carbon nanotubes (COOH-CNTs) to thiolated CNTs, which can be used for covalent coupling with gold nanoparticles or gold-labeled biomolecules.

Purification, Functionalization, and Bioconjugation of Carbon Nanotubes

523

8. Add 100 mL of 88 mM sodium periodate solution immediately to each milliliter of the enzyme solution to give a final periodate concentration of 8 mM in the reaction mixture. 9. Incubate the reaction mixture from Step 8 for 1 h in an ice bath with gentle stirring in the dark. 10. Terminate the reaction by adding either glycerol (0.1  mL glycerol per mL of reaction solution) or ethylene glycol (250  mM) and incubating further for 30  min at room temperature. 11. Purify and concentrate the oxidized enzyme product by dialysis (10 kDa MWCO). 12. Disperse a predetermined amount of the amino-modified CNTs (from Step 6) in PBS buffer (pH 7.5) for 1  h in an ultrasonic water bath. 13. Slowly add the concentrated solution of oxidized glucose oxidase (from Step 11) to the suspension of amino-modified CNTs while gently stirring. Incubate the reaction mixture for 2  h to obtain CNTs covalently conjugated to glucose oxidase. 3.5.4. Covalent Attachment of Biomolecules to Gold Nanoparticle-Decorated CNTs

CNTs decorated with gold nanoparticles (prepared as described in Subheading 3.4) can be conjugated with biomolecules using a variety of different procedures. In the example protocol below, DTSP ­( also known as Lomant’s reagent) adsorbs to the surface of the gold nanoparticles through the disulfide group, so that the terminal succinimidyl groups allow further covalent immobilization of amino-containing biomolecules such as enzymes, ­proteins, etc. (Fig. 4).

Fig. 4.  Adsorption of dithiobis(N-succinimidyl propionate) (DTSP) to gold nanoparticles decorated on CNTs. The adsorbed DTSP can be used to couple the gold/CNT complexes to the free amino groups of biomolecules.

524

Luong et al.

1. Incubate gold nanoparticle-decorated CNTs in 10 mM DTSP in DMSO for 30 min (see Note 15). 2. Wash the resulting nanotubes with DMSO three to five times to remove excess DTSP using a polycarbonate membrane filter (0.22 mm pore size). 3. Wash the functionalized tubes with deionized water three to five times. 4. Incubate the active ester-functionalized gold-decorated CNTs in a buffer solution containing the enzyme of interest (e.g., horseradish peroxidase (6 mg/mL) in 0.1 M phosphate buffer, pH 7) for 1.5–3 h. 5. Rinse the bioconjugated CNTs with buffer three to five times to remove excess enzyme. The bioconjugation of CNTs is currently a subject of intense endeavor; however, the controlled functionalization of CNTs has not yet been fully established. Modification of the CNT surface with metal nanoparticles can potentially impart unique and/or enhanced properties, leading to the production of novel nanomaterials for diversified applications in electrocatalysis, biosensors, biofuel cells, and implantable bioassays.

4. Notes 1. A useful online listing of CNT suppliers may be found at http://www.nanoten.com/ntyp.html. Some specific examples of suppliers include the following: (1) NanoCarbLab (NCL; http://www.nanocarblab.com) offers pristine as well as modified SWCNTs with different purities ranging from 40 to 90% (all percentage purities stated herein are referred to as % w/w values): SWCNT-COOH (70–80% purity, 2–5% of carboxyl (–COOH) groups) and SWCNT-NH2 (70–80% purity, 2–5% of amino (–NH2) groups). Shortened SWCNT-COOH and SWCNT-NH2 (length 200–500 nm) with higher purity (90%) and with a higher content of –COOH or –NH2 groups (5–10%) are also available. The company also offers DWCNTs with 20–30% to 90% purity. As expected, CNTs with a high purity rating are very expensive compared to their counterparts with a low purity rating. (2) NanoLab (http://www. nano-lab.com) is another commercial source of MWCNTs and SWCNTs. The MWCNTs are produced by CVD and purified to >95% as measured by thermal gravimetric analysis (TGA). Two different types of MWCNTs are offered: tubes with lengths of 1–5  mm and diameters of 15 ± 5  nm or 30 ± 10  nm; and longer tubes (5–20  mm) with the same

Purification, Functionalization, and Bioconjugation of Carbon Nanotubes

525

diameters. However, the products still contain 0.94% Fe and 0.14% S compared to 98.92% C. SWCNTs with different purities (40–95%) can also be obtained from NanoLab. SWCNTs with high purity (95%) and shorter lengths (1–5 mm) are prepared by CVD, whereas SWCNTs prepared by arc discharge are longer (over 10 mm) and ­contain 25% Ni or 40–50% Fe. Even after extensive purification, SWCNTs with over 90% purity still contain about 10% Fe. The company also offers DWCNTs (outer diameter = 4 ± 1 nm with purity over 95%) produced by the CVD method. COOH- and NH2functionalized CNTs are also available from this company. Pristine CNTs produced by CVD are ­subject to a reflux in sulfuric/nitric acid to generate a large concentration of carboxyl groups on the CNT surface (2–7% –COOH by titration). Amino-functionalized CNTs are ­produced as a derivative of COOH-functionalized nanotubes, whereby the –COOH group reacts with SOCl2 to form an acyl chloride, which is then reacted with dimethylamine to generate NH2modified CNTs. (3) Carbon Solutions, Inc. (http://www. carbonsolution.com) offers functionalized SWCNTs, such as CNT-PEG (polyethylene glycol), and also supplies AP (asprepared)-SWCNTs with low purity (40–60% carbon and 30% metal content) as well as purified SWCNTs with high purity (>90% carbon and 2–4% metal content). (4) Nanocs (http://www.nanocs.com) offers both SWCNTs (diameters of 2–10  nm, lengths of 50  nm to microns) and MWCNTs (diameters of 10, 20, 40, 60, and 80–150 nm, with lengths of 1–100  mm) with different purities (40% to over 90%). The metal content in such products ranges from 2 to 30%. The company also offers modified CNTs such as CNT-COOH, CNT-NH2, CNT-SH, and CNT-PEG. Such functional groups can be attached either at the ends or the walls of the nanotubes. 2. Both the purity and solubility of prepared CNTs continue to be an issue for practical applications, i.e., new purification and characterization techniques are still needed. For ­example, one purification method developed at the NASA Glenn Research Center can purify gram-scale quantities of SWCNTs. This method uses a combination of high-temperature ­oxidations and repeated extractions with nitric and hydrochloric acid, and is a modification of a gas-phase procedure reported previously by Smalley and coworkers (14). Nanotubes ­purified by this method reveal near complete removal of iron catalyst particles, from 22.7% (w/w) in the crude ­preparation to less than 0.02% (w/w) in the final product. 3. Fine, homogeneous dispersion/suspension of CNTs (1 mg/10 mL solvent) can be prepared using DMF, decalin,

526

Luong et al.

and xylenes. Chloroform is not able to disperse CNTs, whereas chlorobenzene and 1,2-dichlorobenzene can be used to prepare homogeneous dispersions/solutions and homogeneous solutions, respectively. Furthermore, oleum (20% free SO3) can disperse CNTs without sonication; however, such solvents are not compatible for bioconjugation. 4. Sonication of CNTs in 1% (w/v) aqueous SDS results in a solution (50–100 mg CNTs/L) that is stable for about 1 month. However, SWCNTs are less soluble compared to MWCNTs, and only short SWCNTs (150 nm) are able to be dissolved. 5. At each step, the purified products should be examined by standard analytical methods such as TEM, atomic force microscopy (AFM), TGA, Raman spectroscopy, visible-near infrared (NIR) spectroscopy, etc. Unfortunately, sample ­volumes are typically very small and there is no algorithm that can be used to easily convert the images from AFM or TEM into numerical data to quantify purity. When working with small quantities of starting materials, it is always desirable to obtain CNTs with the highest grade of purity from commercial sources (such as those listed as examples in Note 1 above). 6. CNTs are difficult to disperse and cannot be wetted by liquids with surface tensions higher than 100–200  mN/m. Thus, only ~0.1 mg of MWCNTs can be solubilized (with the aid of sonication) in 1  mL of DMF – one of the most effective solvents for CNT solubilization (50). Furthermore, CNTs tend to agglomerate as bundles in solvents, and if dispersed can reagglomerate quickly thereafter due to electrostatic attraction. Nevertheless, wrapping CNTs with a polymer can enhance the solubility of the nanotubes without affecting their original properties (51). CNTs can also be suspended and solubilized in Nafion (52), a perfluorosulfonated polymer, to facilitate the modification of electrode surfaces with an enzyme. Other procedures such as end- (52) and/or sidewall- (53) functionalization, sonication in the presence of a surfactant (54), and protonation by acids (55) are also effective to a certain extent. A concentrated H2SO4/HNO3 mixture can be used to generate carboxylic acid groups at the ends and sidewalls of CNTs via a refluxing/sonication process (56). The resulting oxidized CNTs become more soluble and can be stabilized in aqueous suspensions. Induced defects in the structure of oxidized CNTs can provide sites for the covalent coupling of biomolecules using the standard ­water-soluble coupling agent EDC (57, 58). However, such aggressive oxidative procedures often cut the CNTs into shorter fragments, thus partly reducing their high aspect ratio (59).

Purification, Functionalization, and Bioconjugation of Carbon Nanotubes

527

7. The functionalization of CNTs with biomolecules can be realized by both noncovalent- and covalent-based modification, and several biomolecules can also become adsorbed directly to CNTs noncovalently. In particular, antibodies (60), enzymes (61), and peptides (62) can bind nonspecifically to the exterior surface of CNTs, whereas oligonucleotides can bind to the surface as well as the open cavity space of CNTs. DNA– CNT interactions in water are suggested to result from DNA base-stacking at the CNT surface, with the hydrophilic sugarphosphate backbone oriented outward toward the aqueous medium (63). And finally, ionic liquids and polymers – for instance, 4-chlorobenzenediazonium ­tetrafluoroborate (64) and PDAC (65–71), respectively – are also very efficient in functionalizing CNTs, resulting in charged nanotube structures that may have potential utility in biomedical applications due to their strong interactions with charged cell membranes. Various ionic liquids are able to solubilize CNTs (72–76); the presence of the ionic liquid leads to the exfoliation of SWCNT ropes followed by the addition of functionalized aryl groups to the SWCNT sidewalls, resulting in functionalized SWCNTs that remain ­predominantly as separated strands. 8. In some cases, after immobilization, enzymes may be deactivated or their inherent activity is greatly reduced. One way that may help circumvent this issue is to conduct the immobilization of the target enzyme in the presence of its substrate. 9. Prior to their deposition onto the pyrene-modified substrate, the oxidized SWCNTs are suspended in an aqueous solution and sonicated for 1 h. Following this, the ­suspension is centrifuged at 10,000 × g for 5  min, and then the recovered supernatant is filtered through filter paper (0.45  mm) (29). Alternatively, the SWCNTs used for deposition may be ­stabilized through the addition of highly charged ZrO2 ­nanoparticles (30). 10. Alternatively, capped cobalt or cobalt/platinum nanoparticles may also be used (31). Sonication in concentrated H2SO4 results in a (20 mg/L) CNT suspension that is only stable for a few hours. A very stable CNT suspension of up to 200 mg/L can be obtained by stirring the CNTs in oleum (20% w/w SO3 in H2SO4). 11. As discussed earlier, acid treatment of CNTs at elevated temperature facilitates their dispersion in an aqueous environment. This treatment also creates a variety of carbonyl functional groups at defect sites in the outer graphene sheet. Further treatment with perchloric acid (HClO4) or potassium permanganate (KMnO4) will convert all of these groups into carboxylic acids. The carboxylates, in turn, can be used to link

528

Luong et al.

CNTs to the amino groups of biomolecules or proteins using the well-known carbodiimide procedure (77). In this approach, the reaction between the –N=C=N– moiety of the carbodiimide with the carboxyl group of the CNT forms a highly reactive o-acylisourea derivative with an extremely short half-life (Fig. 2). This active species will react with the amino group of a biomolecule (i.e., enzyme, protein, etc.) to form a stable amide bond. Carbodiimides are “zero-length” crosslinking agents as no additional ­chemical structure is introduced between the CNT and the biomolecule. In some cases, the activity of the immobilized biomolecules may be affected due to steric hindrance induced by the covalent binding process. This carbodiimide coupling procedure is often used to crosslink biomolecules to each other (such as in the synthesis of protein–protein, peptide–protein, nucleotide– protein complexes) as well as to immobilize biomolecules to a substrate. The water-soluble carbodiimide EDC and N-hydroxysulfosuccinimide (sulfo-NHS) are widely used for carrying out this type of coupling reaction. In general, the advantage of adding sulfo-NHS to EDC reactions is that an increase in the stability of the active intermediate is achieved (Fig. 2). In this reaction scheme, nucleophilic attack of the free amino groups of the biomolecule by the activated carboxyl groups on the CNTs results in the CNT–protein conjugate. The protocol ­provided here is generalized and simplified from several different sources that describe in detail methods to couple a protein molecule to CNTs through EDC/sulfo-NHS ­chemistry (78–80). 12. Besides EDC, N-Cyclohexyl-N ′-2(4′-morpholium) ethyl ­carbodiimide-p-toluenesulfonate can also be used for the coupling ­reaction. EDC and N-hydroxysuccinimide (NHS) are widely used together since the NHS produces a more stable reactive ­intermediate, which has been shown to give a greater reaction yield. 13. Carbodiimide coupling reactions using EDC can also be performed in a solution of N-2-hydroxyethylpiperazine-N ¢2-ethanesulfonic acid (HEPES) at an optimal pH of 7.2–7.5. 14. Alternatively, the gold nanoparticle-decorated CNTs can be incubated with 20 mM cystamine (2,2¢-diaminodiethyldisulfide) in deionized water for 2–3 h to introduce –S–(CH2)2–NH2 onto the gold surface. Biomolecules containing –COOH groups can then be coupled to the free amino groups on the gold-decorated CNTs by carbodiimide coupling, similar to the procedure described in Subheading 3.5.2. 15. 1-(3-Aminopropyl)-3-methylimidazolium bromide (ILNH2) may be prepared in the laboratory, as described in (46).

Purification, Functionalization, and Bioconjugation of Carbon Nanotubes

529

References 1. Iijima, S. (1991) Helical microtubules of graphitic carbon. Nature 354, 56–58. 2. Dresselhauss, M.S., Dresselhauss G., and Eklund, P.C. (1996) Science of fullerenes and carbon nanotubes. Academic Press, San Diego, CA, pp. 1–985. 3. Dekker, C. (1999) Carbon nanotubes as molecular quantum wires. Phys. World 52, 22–28. 4. Treacy, M.M.J., Ebbesen, T.W., and Gibson, J.M. (1996) Exceptionally high Young’s modulus observed for individual carbon nanotube. Nature 381, 678–80. 5. Dai, H. (2000) Controlling nanotube growth. Phys World 13, 43–47. 6. Ebbesen, T.W., and Ajayan, P.M. (1992) Large-scale synthesis of carbon nanotubes. Nature 358, 220–222. 7. Ando, Y., and Iijima, S. (1993) Preparation of carbon nanotubes by arc-discharge evaporation. Jpn. J. Appl. Phys. 32, L107–L109. 8. Zhao, X., Wang, M., Ohkohchi, M., and Ando Y. (1996) Morphology of carbon nanotubes prepared by carbon arc. Jpn. J. Appl. Phys. 35, 4451–4456. 9. Iijima, S., and Ichihashi, T. (1993) Single-shell carbon nanotubes of 1-nm diameter. Nature 363, 603–605. 10. Sen, R., Ohtsuka, Y., Ishigaki, T., Kasuya, D., Suzuki, S., Kataura, H, et  al. (2000) Time period for the growth of single-wall carbon nanotubes in the laser ablation process: evidence from gas dynamic studies and time resolved imaging. Chem. Phys. Lett. 332, 467–473. 11. Che, G., Lakshmi, B.B., Martin, C.R., Fisher, E.R., and Ruoff, R.S. (1998) Chemical vapor deposition based synthesis of carbon nanotubes and nanofibers using a template method. Chem. Mater. 10, 260–267. 12. Ebbesen, T.W., Ajayan, P.M., Hiura H., and Tanigaki, K. (1994) Purification of nanotubes. Nature 367, 519–519. 13. Chiang, I.W., Brinson, B.E., Smalley, R.E., Margrave, J.L, Hauge, R.H. (2001) Purification and characterization of single-wall carbon nanotubes. J. Phys .Chem. B, 105, 1157–1161. 14. Chiang, I.W. Brinson, B.E. Huang, A.Y., Willis, P.A., Brownikowski, M.J., Margrave, J.L., et  al. (2001) Purification and characterization of single-wall carbon nanotubes (SWNTs) obtained from the gas-phase decomposition of CO (HiPco process). J. Phys. Chem. B 105, 8297–8301. 15. Haddon, R.C. Sippel, J. Rinzler, A.G., and Papadimitrakopoulos, F. (2004) Purification

and separation of carbon nanotubes. MRS Bull. 29, 252–259. 16. Bandow, S., Rao, A.M., Williams, K.A., Thess, A., Smalley, R.E., and Eklund, P.C. (1997) Purification of single-wall carbon nanotubes by microfiltration. J. Phys. Chem. B 101, 8839–8842. 17. Dujardin, E., Ebbesen,T.W., Krishnan, A., and Treacy, M.M.J. (1998) Purification of singleshell nanotubes. Adv. Mater. 10, 611–613. 18. Dujardin, E. Dny, C.M, Panissod, P. Kintzinger, J.P., Yao, N., and Ebbesen, T.W. (2000) Interstitial metallic residues in purified single shell carbon nanotubes. Solid State Commun. 114, 543–546. 19. Rinzler, A.G., Liu, J., Nikolaev, P., Huffman, C.B., Rodrigues, F.J., Macias, P.J. et al. (1998) Large-scale purification of single-wall carbon nanotubes: process, product, and characterization. Appl. Phys. A 67, 29–37. 20. Dillon, A.C., Gennet, T., Jones, K.M., Alleman, J.L., Parilla, P.A., and Haben, M.J. (1999) A simple and complete purification of singlewalled carbon nanotube materials. Adv.Mater. 11, 1354–1358. 21. Tohji, K., Goto, T., Takahashi, Shinoda, H. Y., Shimizu, N., Jaya-devan, B., et  al. (1996) Purifying single-walled nanotubes. Nature 383, 679–679. 22. Tohji, K., Takahashi, H. Y., Shinoda, N., Shimizu, B., Jayadevan, I., Matsuoka, Y., et al. (1997) Purification procedure for single-walled nanotubes. J.Phys.Chem. B, 101, 1974–1978. 23. Martinez, M.T., Callejas, M.A., Benito, A.M., Maser, W.K., Cochet, M., Andres, J.M., et  al. (2002) Microwave single walled carbon nanotubes purification. Chem. Commun. 1000–1001. 24. Sen, R., Rickard, S.M., Itkis, M.E., and Haddon, R.C. (2003) Controlled purification of single-walled carbon nanotube films by use of selective oxidation and near-IR spectroscopy. Chem. Mater. 15, 4273–4279. 25. Thien-Nga, L., Hernadi, K., Ljubovic, E., Garaj, S., and Forro, L. (2002) Mechanical purification of single-walled carbon nanotube bundles from catalytic particles. Nano Letts. 2, 1349–1352. 26. Bonard, J-M., Stora, T., Salvetat, J-P., Maier, F., Stockli, T., Duschl, C., et al. (1997) Purification and size-selection of carbon nanotubes. Adv. Mater. 9, 827–831. 27. Chen, R.J., Zhang, Y., Wang, D., and Dai, H. (2001) Noncovalent sidewall functionalization of single-walled carbon nanotubes for

530

Luong et al.

protein immobilization. J. Am. Chem. Soc. 123, 3838–3839. 28. Besteman, K., Lee, J-O., Wiertz, F.G.M., Heering, H.A., and Dekker, C. (2003) Enzyme-coated carbon nanotubes as singlemolecule biosensors. Nano Lett. 3, 727–730. 29. Zhu, J., Yudasaka, M., Zhang, M., Kasuya, D. and Iijima, S. (2003) A surface modification approach to the patterned assembly of singlewalled carbon nanomaterials. Nano Letts. 3, 1239–1243. 30. Zhu, J., Yudasaka, M., Zhang, M., and Iijima, S. (2004) Dispersing carbon nanotubes in water: a noncovalent and nonorganic way. J. Phys. Chem. B 108, 11317–11320. 31. Georgakilas, V., Tzitzios, V., Gournis, D., and Petridis, D. (2005) Attachment of magnetic nanoparticles on carbon nanotubes and their soluble derivatives. Chem. Mater. 17, 1613–1617. 32. Liu, J., Rinzler, A. G., Dai, H., Hafner, J. H, Bradley, R. K., Boul, P. J., Lu, A., et  al (1998) Fullerene pipes. Science 280, 5367, 1253–1256. 33. Chen, J., Hamon, M. A., Hu, H., Chen, Y., Rao, A. M., Eklund, P. C., and Haddon, R. C.. (1998) Solution properties of single-walled carbon nanotubes. Science 282, 5386, 95–98. 34. Yan, Y., Zhang, M., Gong, K., Su, L., Guo, Z., and Mao, L. (2005) Adsorption of methylene blue dye onto carbon nanotubes: A route to an electrochemically functional nanostructure and its layer-by-layer assembled nanocomposite. Chem. Mater. 17, 3457–3463. 35. Breuer, O. and Sundararaj, U. (2004) Big returns from small fibers: A review of polymer/ carbon nanotube composites. Poly. Comp. 25, 630–645. 36. Xue, B., Chen, P., Hong, Q., Lin, J. Y., and Tan, K. L. (2001) Growth of Pd, Pt, Ag and Au nanoparticles on carbon nanotubes. J. Mater. Chem. 11, 2378–2381. 37. Li, J., Moskovits, M., and Haslett, T. L. (1998) Nanoscale electroless metal deposition in aligned carbon nanotubes. Chem. Mater.10, 1963–1967. 38. Azamian, B. R., Coleman, K. S., Davis, J. J., Hanson, N., and Green, M. L. H. (2002) Directly observed covalent coupling of quantum dots to single-wall carbon nanotubes. Chem. Commun. 366–367. 39. Kim, B. and Sigmund, W. M. (2004) Functionalized multiwall carbon nanotube/ gold nanoparticle composites. Langmuir 20, 8239–8242.

40. Li. F., Wang, Z., Shan, C., Song, J., Han, D., and Niu, L. (2009) Preparation of gold ­nanoparticles/functionalized multiwalled carbon nanotube nanocomposites and its glucose biosensing application. Biosens. Bioelectron. 24, 1765–1770. 41. Liu, Y., Male, K.B., Bouvrette, P., and J.H.T. Luong. (2003) Control of the size and distribution of gold nanoparticles by unmodified cyclodextrins. Chem. Mat. 15, 4172–4180. 42. Mahmoud, K. A., Hrapovic, S. and Luong, J.H.T. (2008) Picomolar detection of protease using peptide/single walled carbon nanotube/ gold nanoparticle-modified electrode. ACS Nano 2, 1051–1057 43. Mahmoud, K.A. and Luong J.H.T. (2008) Impedance method for detecting HIV-1 ­protease and screening for its inhibitors using ferrocene-peptide conjugate/Au nanoparticle/single-walled carbon nanotube modified electrode. Anal. Chem. 80, 7056–7062. 44. Shie, J.-W., Yogeswaran, U, and Chen, S.-M. (2009) Haemoglobin immobilized on Nafion modified multi-walled carbon nanotubes for O2, H2O2 and CCl3COOH sensors. Talanta, 78, 896–902. 45. Zhang, R. and Wang, X. (2007) One step synthesis of multiwalled carbon nanotube/gold nanocomposites for enhancing electrochemical response. Chem. Mater. 19, 976–978. 46. Li, F., Wang, Z., Shan, C., Song, J., Han, D., and Niu, L. (2009) Preparation of gold nanoparticles / functionalized multiwalled carbon nanotube nanocomposites and its glucose biosensing application. Biosens. Bioelectron. 24, 1765–1770. 47. Hu, X., Wang, T., Qu, X., and Dong, S. (2006) In situ synthesis and characterization of multiwalled carbon nanotube/Au nanoparticle composite materials. J. Phys. Chem. B 110, 853–857. 48. Luong, J.H.T., Hrapovic, S., Wang, D., Bensebaa, F., and Simard, B. (2004) Solubilization of multiwall carbon nanotubes by 3-aminopropyltriethoxysilane towards the fabrication of electrochemical biosensors with promoted electron transfer. Electroanal. 16, 132–139. 49. Li, J., Wang, Y.-B., Qiu, J-D., Sun, D.-C., and Xia, X.-H. (2005) Biocomposites of covalently linked glucose oxidase on carbon nanotubes for glucose biosensor. Anal. Bioanal. Chem. 383, 918–922. 50. Yamamoto, K., Shi G., Zhou, T., Xu, F., Xu, J., Kato, T., et al. (2003) Study of carbon nanotubes-HRP modified electrode and its application for novel on-line biosensors. Analyst 128, 249–254.

Purification, Functionalization, and Bioconjugation of Carbon Nanotubes 51. O’Connell, M.J., Boul, P., Ericson, L.M., Huffman, C., Wang, Y.H., Haroz, E., et  al. (2001) Reversible water-solubilization of singlewalledcarbon nanotubes by polymer wrapping. Chem. Phys. Lett. 342, 265–271. 52. Wang, J., Musameh, M., and Lin, Y. (2003) Solubilization of carbon nanotubes by Nafion toward the preparation of amperometric biosensors. J. Am. Chem. Soc. 125, 2408–2409. 53. Chen, J., Hamon, M.A., Hu, H., Chen, Y., Rao, A.M., Eklund, P.C., et al. (1998) Solution properties of single-walled carbon nanotubes. Science 282, 95–98. 54. Tasis, D., Tagmatarchis, N., Georgakilas, V., and Prato, M. (2003) Soluble carbon nanotubes. Chem. Eur. J. 9, 4000–4008. 55. Islam, M.F., Rojas, E., Bergey, D.M., Johnson, A.T., and Yodh, A.G. (2003) High weight fraction surfactant solubilization of single-wall carbon nanotubes in water. Nano Lett. 3, 269–273. 56. Liu, J., Rinzler, A.G., Dai, H., Hafner, J.H., Bradley, R.K., Boul, P.J., et al. (1998) Fullerene pipes. Science 280, 1253–1256. 57. Hirsch A. (2002) Functionalization of singlewalled carbon nanotubes. Angew. Chem. Int. Ed. 114, 1933–1939. 58. Mawhinney, D.B., Naumenko, V., Kuznetsova, A., Yates Jr,. J.T., Liu, J., and Smalley, R.E. (2000) Surface defect site density on single walled carbon nanotubes by titration. Chem. Phys. Lett. 324, 213–216. 59. Ramesh, S., Ericson, L.M., Davis, V.A., Saini, R.K., Kittrell, C., Pasquali, M., et  al. (2004) Dissolution of pristine single walled carbon nanotubes in superacids by direct protonation. J. Phys. Chem. B 108, 8794–8798. 60. Wang, S., Humphreys, E.S., Chung, S-Y., Delduco, D.F., Lusting, S.R., Wang, H., et al. (2003) Peptides with selective affinity for carbon nanotubes. Nat. Mater. 2, 196–200. 61. Azamian, B.R., Davis, J.J., Coleman, K.S., Bagshaw, C.B., and Green M.L.H. (2002) Bioelectrochemical single-walled carbon nanotubes. J.Am.Chem. Soc. 124, 12664–12665. 62. Dieckmann, G.R., Dalton, A.B., Johnson, P.A., Razal, J., Chen, J., Giordano, G.M., et al. (2003) Controlled assembly of carbon nanotubes by designed amphiphilic peptide helices. J. Am. Chem. Soc. 125, 1770–1777. 63. Zheng, M., Jagota, A., Strano, M.S., Santos, A.P., Barone, P., Chou, S.G., et  al. (2003) Structure-based carbon nanotube sorting by sequence-dependent DNA assembly. Science 302, 1545–1548.

531

64. Price, B.K., Hudson, J.L., and Tour, J.M. (2005) Green chemical functionalization of single-walled carbon nanotubes in ionic liquids. J. Am. Chem. Soc. 127, 14867–14870. 65. Rochette, J.F., Sacher, E., Meunier, M., and Luong, J.H.T. (2005) A mediatorless biosensor for putrescine using multiwalled carbon nanotubes. Anal. Biochem. 336, 305–311. 66. Zhao, H., and Ju, H. (2006) Multilayer membranes for glucose biosensing via layer-by-layer assembly of multiwall carbon nanotubes and glucose oxidase. Anal. Biochem. 350, 138–144. 67. Zhang, M., Gong, K., Zhang, H., and Mao, L. (2005) Layer-by layer assembled carbon nanotubes for selective determination of dopamine in the presence of ascorbic acid. Biosens. Bioelectron. 20, 1270–1276. 68. Liu, G., and Lin, Y. (2006) Carbon nanotubetemplated assembly of protein. J. Nanosci. Nanotechnol. 6, 948–953. 69. Yang, D-Q., Rochette, J-F., and Sacher, E. (2005) Spectroscopic evidence for p-p interaction between poly(diallyl dimethylammonium) chloride and multiwalled carbon nanotubes. J. Phys. Chem. B. 109, 4481–4484. 70. He, P., and Bayachou, M. (2005) Layer-bylayer fabrication and characterization of DNA wrapped single-walled carbon nanotube particles. Langmuir. 21, 6086–6092. 71. Liu, G., and Lin, Y. (2006) Biosensor based on self-assembling acetylcholinesterase on carbonnanotubes for flow injection/amperometric detection of organophosphate pesticidesand nerve agents. Anal. Chem. 78, 835-843. 72. Barisci, J.N., Wallace, G.G., MacFarlane, D.R., and Baughman, R.H. (2004) Investi-gation of ionic liquids as electrolytes for carbon nanotube electrodes. Electrochem. Comm. 6, 22–27. 73. Katakabe, T., Kaneko, T., Watanabe, M., Hukushima, T., and Aida, T. (2005) Electric double layer capacitors using ionic liquids and carbon nanotubes. J. Electrochem. Soc. 152, A1913–A1916. 74. Fukushima T., Kosaka, A., Ishimura Y., Yamamoto T., Takigawa T., Ishii N., and Aida T. (2003) Molecular ordering of organic molten salts triggered by single-walled carbon nanotubes. Science 300, 2072–2074. 75. Yu, B., Zhou, F., Liu, G., Liang, Y., Huck, W.T.S., and Liu, W. (2006) The electrolyte switchable solubility of multi-walled carbon naotube/ionic liquid (MWCNT/IL) hybrids. Chem. Comm. 2356–2358. 76. Arapalli, S., Fireman, H., Huffma, C., Moloney, P., Nikolaev, P., Yowell, et  al

532

Luong et al.

(2005) Carbon-nanotube-based electrochemical double-layer capacitor technologies for spaceflight applications. JOM 57, 26–31. 77. Hermanson G. T. (2008). Bioconjugate Techniques, 2nd Edition. Elsevier/Academic Press, Amsterdam-San Diego, CA. 78. Anjaneyulu, P.S., and Staros, J.V. (1987) Reactions of N-hydroxysulfosuccinimide active esters. Int. J. Pept. Protein Res. 30 (1), 117–124.

79. Thelen, P., and Deuticke, B. (1988) Chemomechanical leak formation in human erythrocytes upon exposure to a water-soluble carbodiimide followed by very mild shear stress. II. Chemical modifications involved. Biochim Biophys Acta 944, 297–307. 80. Grabarek, Z. , and Gergely, J. (1990) Zerolength crosslinking procedure with the use of active esters. Anal. Biochem. 185, 131–135.

Chapter 33 Functional Integration of Membrane Proteins with Nanotube and Nanowire Transistor Devices Aleksandr Noy, Alexander B. Artyukhin, Shih-Chieh Huang, Julio A. Martinez, and Nipun Misra Abstract Biological molecules perform a sophisticated array of transport and signaling functions that rival anything that the modern electronics industry can create. Incorporating such building blocks into nanoelectronic devices could enable new generations of electronic circuits that use biomimetics to perform complicated tasks. Such types of circuits could ultimately blur the interface between living biological organisms and synthetic structures. Our laboratory has recently developed a versatile and flexible platform for integrating ion channels and pumps into single-walled carbon nanotube (SWNT) and silicon nanowire (SiNW) transistor devices, in which membrane proteins are embedded in a lipid bilayer shell covering the nanotube or nanowire component. In this chapter, we provide details for the fabrication of these devices and outline procedures for incorporating biological molecules into them. In addition, we also provide several examples of the use of these devices to couple biological transport to electronic signaling. Key words: Membrane proteins, Ion channels, Carbon nanotube transistor, Silicon nanowire transistor, Lipid bilayer, Bionanoelectronics

1. Introduction Nature has evolved a vast arsenal of highly specific receptors, active and passive ion channels (1, 2), and photoactivated ion (proton) pumps (3) to achieve effective control over information flow from the environment to the cell. Using these components in electronic circuits could lead to the seamless integration of biological and synthetic structures (4), the development of superior biosensing and diagnostics tools (5), advanced neuroprosthetics (6), and perhaps even more efficient computers (7). Specific examples of the integration of biological systems with microelectronic devices include earlier studies on the capacitive Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_33, © Springer Science+Business Media, LLC 2011

533

534

Noy et al.

stimulation of cells (8) and the monitoring of neuronal activity with field-effect transistors (9), as well as a more recent demonstration of the use of silicon nanowire (SiNW) transistor arrays to follow neuronal signal propagation (6). Nanomaterials that have characteristic dimensions comparable to the size of biological molecules enable a much more localized implementation of such integration strategies. For example, single-walled carbon nanotubes (SWNTs) can serve as versatile carriers for transporting intracellular proteins and DNA (10), and silicon nanowires (SiNWs) can be employed as efficient gene delivery vehicles for mammalian cells (11). And furthermore, researchers have also taken advantage of the superior electronic properties of silicon nanowires to create highly specific, label-free electronic detectors for a variety of biomolecules (5). Building bionanoelectronic interfaces requires the functional integration of nanomaterials with membrane-bound proteins. Lipid membranes occupy a special place within the hierarchy of cellular structures as they represent an important structural and protective component of the cell that forms a stable, self-healing, and virtually impenetrable barrier to ions and hydrophilic molecules (12). A lipid membrane also represents a nearly universal matrix that can host a virtually unlimited number and variety of protein nanomachines (13–16). As described in this chapter, we have integrated lipid bilayer membranes with SWNT and SiNW transistors by covering the nanostructured devices with a continuous lipid bilayer shell that forms a barrier between the nanowire/ nanotube surface and the various species in solution. We further show that when this “shielded wire” structure incorporates transmembrane protein pores, it enables the conversion of ionic currents to electronic signals via specific ion transport through the integrated membrane pores.

2. Materials 2.1. Single-Walled Carbon Nanotube Device Fabrication

1. Silicon wafers (100), 4-in. diameter (International Wafer Service). 2. Mask alignment system (EVG620) (EV Group Inc.). 3. Photoresists: AZ 5214-E, AZ 1518 (Clariant). 4. Nonphotosensitive resist: LOR-3A (Microchem). 5. AZ developer (Clariant). 6. PRS-2000 stripper (Mallinckrodt Baker). 7. Cr, Pt, Al, Fe, and Mo targets for electron-beam (e-beam) deposition (Plasmaterials).

Functional Integration of Membrane Proteins with Nanotube

535

Fig. 1. Mask layout for fabricating carbon nanotube devices.

8. Three Cr/glass masks for lithographic patterning of electrodes (Layer 1), nanotube catalyst (Layer 2), and contact insulation (Layer 3). We have found that optimal results are achieved when the gap size between the electrodes, which defines the length of the nanotube channel, is 5 mm; the catalyst islands are placed within 1  mm from the electrode edges; and the insulation layer has a 2-mm opening in the center of the nanotube (Fig. 1). 9. Compressed gases for chemical vapor deposition (CVD) synthesis of carbon nanotubes: Ethylene (C2H4), research purity (Matheson Gas Products); hydrogen (H2), research purity (Airgas); and argon (Ar), ultrahigh purity (Air Liquid America Corp). 10. Tube furnace, 1-in diameter (Lindberg Blue). 11. Hydrogen peroxide, aqueous solution (30%, w/w); 12. Sulfuric acid (96%, w/w). 13. 1-mm diameter 3 M KCl Ag/AgCl reference microelectrode (Warner Instruments) or 0.25-mm diameter gold wire (Sigma–Aldrich). 14. Signatone S-1160 Corporation). 2.2. Chemical Vapor Deposition Synthesis of Silicon Nanowires

Probe

Station

(Lucas

Signatone

1. Silicon wafers (100), 4-in. diameter (International Wafer Service). 2. Tube furnace, 1-in diameter (Lindberg Blue). 3. Acetone.

536

Noy et al.

4. Propanol-2 (IPA). 5. Compressed gases for CVD synthesis of silicon nanowires: 10% (v/v) silane (SiH4) in helium (He) (Voltaix); 100 ppm diborane (B2H6) in He (Voltaix); and Ar (Airgas). 6. 0.1% (w/v) Poly-l-lysine, aqueous solution (Ted Pella). 7. Colloidal gold suspensions (Ted Pella): 20- and 30-nm nanoparticle sizes. 8. Ethanol, 200 proof (VWR). 2.3. Silicon Nanowire Transistor Device Fabrication and Characterization

1. Silicon wafers (100), 4-in. diameter, p-type, with a resistivity of 1 MW cm (International Wafer Service). 2. Mask alignment system (EVG620) (EV Group Inc.). 3. Photoresists: AZ 5214-E, AZ 1518 (Clariant). 4. Nonphotosensitive resist: LOR-3A (Microchem). 5. AZ developer (Clariant). 6. AZ 300 MIF developer (Clariant). 7. PRS-2000 stripper (Mallinckrodt Baker). 8. Remover PG stripper (Microchem). 9. Hexamethyldisiloxane (HMDS) (Clariant). 10. Ti, Au, Pt, and Ni targets for e-beam deposition (Plasmaterials). 11. Two Cr/glass masks for lithographic patterning of electrodes: The first mask defines the outer contact pads for electrical probing of the devices. The second mask defines the electrical interconnects between the silicon nanowires and the outer contact pads. 12. Compressed gases for SiNW annealing: H2, research purity (Airgas) and nitrogen (N2). 13. Plasma-enhanced chemical vapor deposition (PECVD) system (SEMI Group 1000 MPB). 14. Compressed gases for silicon nitride PECVD: SiH4 and ammonia (NH3) (Airgas). 15. Buffered oxide etch (HF) with surfactant. 16. Polydimethylsiloxane (PDMS, Sylgard 184) stamp with a microchannel and inlet/outlet holes for fluidic connections. 17. 1-mm diameter 3 M KCl Ag/AgCl reference microelectrode (Warner Instruments). 18. Keithley Model 2602 SourceMeter instrument. 19. Stanford Research SR8650 lock-in amplifier.

Functional Integration of Membrane Proteins with Nanotube

2.4. Cyclic Voltammetry on Nanowire and Nanotube Electrodes

2.5. Lipid Bilayer Assembly and Sodium Pump Reconstitution on Carbon Nanotube Devices

537

1. 1-mm diameter 3 M KCl Ag/AgCl reference microelectrode (Warner Instruments). 2. Potassium ferrocyanide (K4[Fe(CN)6]), ³99.9% purity. 3. K2HPO4, KH2PO4, KCl, HCl, and KOH; all with ³99.9% purity. 1. PDMS (Sylgard 184) stamp with a microchannel and inlet/ outlet holes for fluidic connections. 2. Octaethylene glycol monododecyl ether (C12E8) (Sigma). 3. Bio-Beads SM-2 polystyrene adsorbent beads (20–50 mesh) (Bio-Rad Laboratories). 4. 1,2-Diphytanoyl-sn-glycero-3-phosphocholine (DPhPC), 10 mg/ml in chloroform (Avanti Polar Lipids, Inc.). 5. 1,2-Diphytanoyl-sn-glycero-3-phosphoethanolamine (DPhPE), 10 mg/ml in chloroform (Avanti Polar Lipids, Inc.) 6. 1,2-Dioleoyl-sn-glycero-3-phosphoinositol ammonium salt (PI) (Avanti Polar Lipids, Inc.). 7. Cholesterol (Supelco). 8. Na+,K+-ATPase (Sigma). 9. ATPase buffer: 130 mM NaCl, 20 mM KCl, 4 mM MgCl2, and 30 mM histidine (pH 7.5). 10. Water bath sonicator (Cole-Parmer 8891 ultrasonic cleaner).

2.6. Lipid Bilayer Assembly on Silicon Nanowire Devices

1. 1,2-Dioleoyl-sn-glycero-3-phosphocholine (DOPC), 10 mg/ml in chloroform (Avanti Polar Lipids, Inc.). 2. 1-Oleoyl-2-[6-[(7-nitro-2-1,3-benzoxadiazol-4-yl)amino] hexanoyl]-sn-glycero-3-phosphoethanolamine (NBDPE), 1 mg/ml in chloroform (Avanti Polar Lipids, Inc.). 3. Ethanol, 200 proof (VWR). 4. K-PBS buffer: 100  mM KCl, 5  mM potassium phosphate (K2HPO4/KH2PO4 molar ratio = 3/2), pH 7. Prepare the buffer solution with degassed DI water (see Note 1) and, if necessary, adjust pH to 7 with HCl or KOH.

2.7. Formation of a-Hemolysin Channels in Lipid Bilayers Assembled on Silicon Nanowire Devices

1. a-Hemolysin from Staphylococcus aureus (Sigma–Aldrich).

538

Noy et al.

2.8. Formation of Alamethicin Channels in Lipid Bilayers Assembled on Silicon Nanowire Devices 2.9. Formation of Gramicidin Channels in Lipid Bilayers Assembled on Silicon Nanowire Devices

1. Alamethicin from Trichoderma viride, ³90% (HPLC) (Sigma–Aldrich).

1. Gramicidin A from Bacillus brevis, ³90% (HPLC) (BioChemika). 2. Ethanol, 200 proof (VWR).

3. Methods 3.1. Fabrication of Single-Walled Carbon Nanotube Devices

1. Oxidize 4-in. (100) Si wafers in a dry oxygen (O2) oven at 1,200°C for 4 h. The final oxide thickness should be 400 nm (Fig. 2). 2. Coat the oxidized silicon wafers with AZ 5214-E image reversal photoresist (see Note 2) and pattern it for electrode deposition as follows: dehydrate the wafers in a convection oven at 120°C for at least 30 min unless they are taken straight from the

Fig. 2. Fabrication and characterization of single-walled carbon nanotube transistor devices. (a) Schematic of the nanotube transistor fabrication process: (I) Patterning of metal electrodes on a SiO2 /Si wafer; (II) nanotube catalyst deposition; and (III) carbon nanotube growth by catalytic CVD. (b) SEM image of the final nanotube transistor device. An arrow points to an individual single-walled carbon nanotube (SWNT) bridging two metal electrodes. (c) Transfer characteristics of a representative SWNT device (measured using a source–drain voltage of 100 mV) in air with a back gate (o) and in 100 mM NaCl in water with a liquid gate (). An SEM image of the device is shown in the inset (adapted from (17)).

Functional Integration of Membrane Proteins with Nanotube

539

oxidation furnace (Step 1). Spin coat AZ 5214-E ­photoresist on the wafer at room temperature at 3,000  rpm, and then prebake the wafer in a convection oven at 90°C for 15 min. Expose the wafer through the electrode pattern mask (Layer 1) in the contact aligner tool (EVG620) (exposure time = 1.0 s using a UV light power level of 20 mW/cm2). Since this is the first layer to be patterned, no wafer-to-mask alignment is required at this step. Make sure, however, to align the rectangular edges of the chip pattern with the flats of the wafer to facilitate subsequent dicing of the wafer (Step 5). Next, bake the wafer on a contact hotplate at 115°C for 2 min (see Note 3), and then flood-expose the wafer with UV light for 120 s (no mask). After flood-exposure, develop the wafer in freshly prepared AZ developer (diluted 1:1 v/v with deionized (DI) water) for 30 s. Wash the developed wafer thoroughly with DI water and blow dry. Finally, descum the wafer in a barrel etcher for 1.3 min (1 Torr, 300 W plasma, 300 sccm O2) to remove residual traces of photoresist. 3. E-beam deposit 5 nm Cr/50 nm Pt onto the patterned wafer using a deposition rate of 1–2 Å/s. The chamber pressure during deposition should not exceed (1 − 3) × 10−6 Torr. After metal deposition, lift-off the photoresist layer in hot (60–80°C) PRS2000 stripper. The lift-off process can take up to 1 h to complete. Wash the wafer thoroughly with DI water and blow dry. 4. Coat the wafer with AZ 1518 positive photoresist and pattern it for nanotube catalyst deposition as follows: Dehydrate the wafer in a convection oven at 120°C for at least 30 min. Spin coat AZ 1518 photoresist onto the wafer at room temperature at 2,500 rpm, and then prebake the wafer in a convection oven at 90°C for 15  min. Align the wafer with the nanotube catalyst mask (Layer 2) so that the catalyst islands are positioned symmetrically near the edges of the electrodes. Expose the wafer through the catalyst island pattern mask in the contact aligner tool (exposure time = 12.5 s; power = 20 mW/cm2). Develop the exposed wafer in freshly prepared AZ developer (diluted 1:1 v/v with DI water) for 40 s. Wash thoroughly with DI water and blow dry. 5. Dice the patterned wafer into individual chips using a diamond scriber. 6. E-beam deposit 15 nm Al/2 Å Mo/5 Å Fe catalyst onto a patterned wafer chip (see Note 4). The system pressure should not exceed 1 × 10−6  Torr during the deposition process. Perform e-beam deposition in manual mode to achieve more accurate rate and thickness control. The metal deposition rates should be 50+ Å/s for Al (see Note 5) and 0.02–0.05 Å/s for Mo and Fe. The aluminum target should be in an intermetallic crucible. For the Mo and Fe depositions, use 10 or 20% of the bulk density value and 10× or 5× the nominal thickness value,

540

Noy et al.

respectively. After deposition, lift-off the photoresist layer in hot (60°C) PRS-2000 stripper. The lift-off process takes 5–10 min. Wash thoroughly with DI water and blow dry. 7. Grow single-walled carbon nanotubes (SWNTs) by using an ambient pressure ethylene CVD process as follows: Place a catalyst-patterned chip (ca. 2 × 2 cm) on a half-pipe boat and insert the boat into a 1-in tube furnace (Lindberg Blue). Ramp the furnace temperature to 850°C at 40°C/min while flowing 900 sccm Ar and 400 sccm H2. Five minutes after the furnace has reached 850°C, turn on 3 sccm C2H4 for 10 min. Simultaneously turn on Ar/H2O flow (5–15 sccm, with the argon gas passing through a water bubbler) to keep the water concentration in the gas mixture around 200  ppm. After 10 min of C2H4 flow (15 min total time at 850°C) turn off H2, H2O, and C2H4 and air-cool the furnace down to room temperature while maintaining the Ar flow. 8. After nanotube growth, remove amorphous carbon deposits from the device chip by submerging it in cold piranha solution (3 vol. of 96% w/w H2SO4 + 1 vol. of 30% w/w H2O2) for 3 min. Caution: Piranha solution reacts violently with most organic materials and must be handled with extreme care! Wash the chips thoroughly with DI water and blow dry. 9. Cover the electrodes with a protective layer of LOR-3A nonphotosensitive resist while exposing the central region of the nanotube as follows: Dehydrate the wafer in a convection oven at 125°C for 1 h. Spin coat LOR-3A at room temperature at 4,500  rpm for 30  s to obtain a 300-nm thick film. Hard-bake the wafer on a hotplate at 180°C for 5 min, and then spin coat AZ 5214-E photoresist onto the wafer at 4,500 rpm. Bake the wafer in a convection oven at 90°C for 15  min. Align the wafer with the nanotube insulation mask (Layer 3). Expose the wafer through the nanotube insulation pattern mask in the contact aligner tool (exposure time = 4  s; power = 15  mW/cm2). Bake the wafer on a hotplate at 120°C for 90  s. Flood-expose the wafer with UV light for 130 s (no mask) and develop in AZ 300 MIF developer for 150 s. Rinse the wafer with DI water and blow dry. Remove any remaining AZ 5214-E photoresist by rinsing the wafer in acetone (see Note 6). 10. Attach a PDMS stamp with a 500 × 100  mm (w × h) microchannel to the device chip as follows: Measure and thoroughly mix the PDMS prepolymer solution and the curing agent at a ratio of 10:1 (w/w). Set aside for approximately 45 min to remove air bubbles. Sonicate a prefabricated PDMS stamp in ethanol for 5 min. Spin the wet PDMS mixture at 6,000 rpm for 4  min on a cover slip. Place the PDMS stamp on the PDMS-coated cover slip until the entire interface wets (this may take several minutes; do not apply pressure to the stamp

Functional Integration of Membrane Proteins with Nanotube

541

at this time). Detach the stamp from the cover slip and place it on the nanotube device chip while manually aligning the middle of the PDMS channel with the gaps between the metal electrodes on the device chip. Bake the chip + stamp assembly in a 75°C oven for 30 min. 11. To screen for successfully connected SWNT devices displaying semiconductor behavior (with high gate voltage sensitivity), place the chip on a conductive chuck in a shielded probe station (Signatone S-1160) and connect the probes to the source and drain contact pads on the device chip. Use the conductive chuck to apply a back gate voltage to the device. For characterization in air, apply a 100-mV source–drain bias with a sourcemeter (Keithley 2602) and sweep the gate voltage from −10 to 10 V at 10 V/s while recording the source–drain current. For measurements in fluid, insert a 1-mm Ag/AgCl reference microelectrode into an outlet connected to the PDMS channel and use it to set the gate voltage (Fig. 2c). (Alternatively, a thin gold wire may be used instead of a Ag/AgCl microelectrode.) Use of a reference electrode reduces artificial responses commonly seen with conventional electrode materials. The gate sweep range for liquid gating is −0.5/0.5 V at 10 mV/s. 3.2. CVD Synthesis of Silicon Nanowires

1. Dice oxidized silicon wafers into small pieces (0.5 × 0.5 cm) to use as growth substrates. Clean the wafer pieces by sonication in acetone and then in IPA for 15 min each. Blow the wafer pieces dry with N2 after sonication. 2. Functionalize the wafer substrate by covering with a few drops of 0.1% (w/v) poly-l-lysine for 3 min. Rinse the polyl-lysine solution off with DI water for 10 s and dry with a N2 stream. 3. Place a drop of 20-nm or 30-nm gold colloid solution onto the functionalized substrate for 15 s, rinse with DI water for 10  s, and dry with N2. Colloidal gold nanoparticles of different sizes may be used to obtain the desired nanowire diameters (18). 4. After gold catalyst deposition, clean the substrates in an O2 plasma at 100 W for 5 min. 5. Place the substrate in a quartz boat and insert the boat into a 1-in. tube furnace (Lindberg Blue). Evacuate the CVD system to its base pressure (<50  mTorr) and heat to 440°C. The temperature should be ramped as fast as possible to reach the growth temperature. Once the temperature stabilizes, set the pressure of the chamber to 100 Torr under 35 sccm of Ar flow. Turn off the Ar flow, introduce 30 sccm of SiH4 (10% v/v in He) and 4 sccm of B2H6 (100 ppm in He), and flow the gases for 20  min at a chamber pressure of 100  Torr (see Note 7). The resulting nanowires should be several

542

Noy et al.

Fig. 3. Characterization of silicon nanowire transistor devices. (a) SEM image showing the channel region of a transistor device where an individual silicon nanowire (SiNW) bridges two metal electrodes. The inset shows a photograph of the chip covered with a PDMS microchannel stamp. (b) TEM image of a chemical vapor deposition (CVD)-grown silicon nanowire. (c) Transfer characteristics of a representative SiNW transistor device measured in buffer solution (adapted from (19)).

micrometers long and 20–30  nm in diameter, with smooth surfaces when inspected by scanning electron microscopy (SEM)/transmission electron microscopy (TEM) (Fig. 3b). 6. After evacuating and cooling the tube furnace, remove the wafer chip containing the grown silicon nanowires. To obtain a suspension of the nanowires for deposition during device fabrication, place the chip with the grown nanowires in 1 ml of ethanol and sonicate it in a water bath sonicator for 3–5 s (see Note 8). Remove the chip and keep the ethanol suspension of dispersed nanowires. 3.3. Silicon Nanowire Device Fabrication and Characterization (Fig. 3)

1. Oxidize 4-in., p-type (1 0 0) silicon wafers (1–10 MW cm) to produce an insulated substrate. To form a 250-nm thick oxide layer, oxidize the wafers in a dry O2 furnace at 1,200°C for 2.5  h. If the substrate is to be used as a back-gate for the nanowire transistors, we recommend using lower resistivity silicon wafers and/or smaller oxide thicknesses. 2. Dehydrate the oxidized silicon wafers at 125°C for 2 h, and then immediately prime them by exposing to HMDS vapor for 2 min. It is important to ensure that the wafer surface is still at high temperature during this step, so expose the wafers to a HMDS vapor immediately after taking them out of the dehydration oven. 3. Coat the wafers with a photoresist layer and pattern the outer contact pads as follows: Deposit AZ 5214-E photoresist onto the wafer by spin coating at 500  rpm for 10  s and then at 3,500 rpm for 30 s at room temperature. The resulting thickness of the photoresist layer should be ~1.4  mm. Bake the photoresist in a convection oven for 15 min at 90°C. Expose the wafer in a contact aligner tool for 3 s at a UV light intensity of 20 W/m2 through the mask to pattern the outer contact pads. No precise alignment is required at this step, although

Functional Integration of Membrane Proteins with Nanotube

543

the wafer flats should be oriented with respect to the mask features to make the subsequent dicing of the wafer into chips (Step 8) easier to perform. 4. After exposure, hard-bake the wafer at 120°C for 90 s on a hot plate. The optimum temperature for this step may need to be experimentally determined, but usually ranges between 115°C and 125°C (see Note 9). Flood-expose the wafer to UV light at an intensity of 14 mW/cm2 for 140 s. Develop the wafer by immersion in AZ developer solution diluted 1:1 (v/v) with DI water for 40  s with gentle agitation, followed by thorough rinsing with DI water. Blow dry the wafer with N2. 5. Descum the patterned wafers in a barrel etcher for 1  min (1 Torr, 300 W plasma, 300 sccm O2). 6. E-beam deposit 20  nm Ti/80  nm Au onto the patterned wafer. The chamber pressure during deposition should not exceed (2 − 3) × 10−7 Torr and the deposition rate should be 1–2 Å/s. 7. Lift-off the photoresist layer in hot (80–90°C) PRS-2000 stripper. The lift-off process can take up to 2 h to complete. Wash the patterned wafer thoroughly with DI water and blow dry. Inspect the metal patterns under a microscope at this stage: the contact pads and electrodes should have smooth, straight edges and sharp corners, and should be devoid of any remaining photoresist. The metal patterns should also be continuous and devoid of any cracks and voids. 8. The wafer can now be diced into individual chips for the fabrication of the SiNW devices. 9. Dehydrate the chips at 120°C for at least 30 min in a convection oven. Functionalize the regions of the chip surface that will contain the nanowire devices by coating them with drops of a 0.1% (w/v) aqueous solution of poly-l-lysine for 1 min, rinsing with DI-water, and drying with N2. 10. Thoroughly mix the PDMS prepolymer and curing agent together at a ratio of 10:1 (w/w). Set aside to let air bubbles escape. Pour the PDMS mixture onto a prefabricated master pattern for creating the microchannels and bake at 150°C for 20 min. The size of the individual flow channels should be 500 mm × 50 mm × 1 cm (w × h × l). Cut the PDMS mold into individual channel “stamps.” Sonicate a fabricated PDMS stamp in ethanol for 5 min, and blow-dry with a N2 stream. Press the PDMS stamp containing a microchannel onto the chip and flow the ethanol suspension of silicon nanowires (obtained from Subheading  3.2) onto the poly-l-lysinefunctionalized substrate surface. The nanowires will tend to become aligned with the liquid flow direction; to obtain the highest degree of alignment, the optimum flow velocity of the

544

Noy et al.

nanowire suspension should be adjusted to ~0.1 ml/min. After nanowire deposition, dry off any residual solution in the channel and remove the PDMS stamp. Inspect the chip under a dark-field microscope to determine if the density of nanowires on the surface is sufficient; if it is not, adjust the concentration of the nanowire suspension and/or the total flow time. 11. Use a low-power O2 plasma cleaning step (300 W, 300 sccm O2) for 10 min to remove any residual organic material from the wafer chips, and then anneal the nanowires at 200°C for 10 min in forming gas (10% v/v H2 in N2) to promote adhesion of the nanowires to the substrate. 12. Dehydrate the wafer chip in a convection oven at 125°C for 2 h and allow it to cool down to room temperature. Spin coat LOR-3A at room temperature for 10 s at 500 rpm and then for 30 s at 4,500 rpm to obtain a ~350-nm thick film. Hardbake the chip on a hotplate at 180°C for 5 min. Make sure to wipe off any excess photoresist on the backside of the chip between the coating and baking steps. 13. After cooling the wafer chip to room temperature, spin coat AZ 1518 photoresist onto the chip at 500  rpm for 10  s followed by 4,500  rpm for 30  s. The thickness of the AZ 1518 photoresist layer should be ~1 mm. Postbake the wafer chip for 90 s at 110°C for 50 s. Make sure to wipe off any excess photoresist on the backside of the chip between the coating and baking steps. 14. Align the wafer chip with the second layer mask and expose the chip in an aligner tool in hard contact mode. The second layer mask is used to define the metal interconnects between the previously formed outer contact pads that are used to probe the devices and the metal contacts to the silicon nanowires. For this patterning step, the UV light exposure time is 9  s at a power level of 20  mW/cm2. Develop the exposed wafer in freshly prepared AZ-300 MIF for 1 min with gentle agitation. Wash thoroughly with DI water and blow dry. 15. Descum the chip in a barrel etcher for 2 min (300 W, 1 Torr, 300  sccm O2). Etch away the native oxide on the silicon nanowires in the open contact regions for 10 s in a buffered oxide etch (HF) with surfactant, followed by rinsing for 20 s with DI water and drying with N2 for 15 s. 16. E-beam deposit 10 nm Ti/80 nm Pt or 70 nm Ni onto the patterned chip. The chamber pressure during deposition should not exceed (2 − 3) × 10−7 Torr. 17. Deposit a conformal layer of silicon nitride to passivate the metal contacts as follows: Deposit 80  nm of stoichiometric silicon nitride in a PECVD system at 100°C, 300 mTorr, and

Functional Integration of Membrane Proteins with Nanotube

545

30 W of plasma power using 7.2 sccm SiH4 and 73 sccm NH3 as the source gases (see Notes 10 and 11). 18. Lift-off the photoresist layer in a 90°C bath of Remover PG. The lift-off process with the silicon nitride layer can take up to 4 h to complete (see Note 12). Wash the wafer chip thoroughly with DI water and blow dry. 19. After lift-off, anneal the device chips further in forming gas at 450°C for 3  min to ensure good SiNW/metal contacts. For Ni contacts, the annealing temperature should be 380°C (see Note 13). The temperature ramp-up rate during the annealing step is 100°C/min. Use a temperature soak at 200°C for 2 min to eliminate any moisture prior to carrying out the higher temperature contact anneal. 20. For devices to be used in electrochemical measurements (Subheading 3.4), deposit an additional layer of photoresist to prevent leakage from the metal contact layers as follows: Spin coat the SiNW device chip with a 400-nm thick layer of LOR-3A and bake at 180°C for 5 min. Following this, spin coat the chip again with a layer of photoresist. Pattern rectangular openings 5–7  mm wide × 30  mm long by photolithography to expose the nanowire surfaces, but not the nitride-coated metal contacts. Finally, rinse the substrate with acetone to ensure that the photoresist is completely removed and that the LOR-3A layer remains on the surface. The resulting net dielectric stack of 400  nm LOR-3A and 80  nm PECVD silicon nitride ensures minimal leakage during voltammetric characterization. 21. Inspect the devices with an optical dark-field microscope or by SEM to identify devices with nanowires bridging the source–drain electrodes. SEM inspection should also be used to ensure that a suitably long region of the nanowire channel is not covered by the silicon nitride passivation layer; otherwise, the undercutting process in the photolithography step should be optimized. 22. Attach a PDMS stamp with a 500 × 100 mm (w × h) channel to the dice chip. Measure and thoroughly mix the PDMS prepolymer solution and curing agent at a ratio of 10:1 (w/w). Set aside the PDMS mixture for approximately 45  min to remove air bubbles. Sonicate a prefabricated PDMS stamp in ethanol for 5 min. Spin the wet PDMS mixture at 6,000 rpm for 4 min on a cover slip. Place the PDMS stamp onto the cover slip until the entire interface wets (this step may take several minutes; do not apply pressure to the stamp at this time). Detach the stamp and place it onto the device chip. Bake the chip + stamp assembly in a 75°C oven for 30 min to form a strong bond between the chip surface and the PDMS microchannel stamp.

546

Noy et al.

23. Mount the SiNW device chip into a probe station and test the performance of the nanowire transistor in a DI water environment, with the solution gate voltage swept between −0.5 and +0.5 V (sweep rate = 10 mV/s) using a Ag/AgCl reference microelectrode. In our laboratory, DC device measurements were performed with a Keithley Model 2602 SourceMeter instrument, and AC measurements were performed with a Stanford Research SR8650 lock-in amplifier at frequencies of 19 or 23 Hz. The current–voltage (I–V) curves of the nanowire transistors should be linear at low source–drain bias voltages if the contacts are ohmic. The measurements should preferably be performed in a shielded probe station to minimize interference from electrical noise. 3.4. Cyclic Voltammetry on Silicon Nanowire and Carbon Nanotube Electrodes

1. Cyclic voltammetry (CV) is typically performed on devices with a single metallic contact to a single nanowire/nanotube, although other configurations may also be used. For these measurements, we recommend depositing an additional passivation layer of LOR-3A as described in the preceding device fabrication protocols (see Subheadings 3.1 and 3.3) to avoid any leakage currents from the relatively large metal contact surfaces. 2. Either a PDMS microchannel or a microreservoir may be used for these measurements by attaching it to the device chip as described in the preceding protocols (see Subhea-ding  3.3, Step 22). Fill the PDMS microchannel or reservoir with the appropriate solution for the CV measurements and then insert a Ag/AgCl reference microelectrode into the liquid. 3. CV measurements are performed on the SiNW/CNT devices using a two-electrode setup. In this approach, a nanowire or nanotube acts as the working electrode and the potentials are swept against the reference electrode in steps of 5 mV for 1 s or 6 mV for 3 s from 0 to +0.6 V with a source-meter (Keithley 2602). A solution containing 100 mM KCl in 5 mM potassium phosphate buffer (pH 7) was used to determine the capacitive currents, whereas a 5-mM potassium phosphate buffer (pH 7) containing either 10 mM K4[Fe(CN)6] (for a multiple-nanowire electrode) or 50 mM K4[Fe(CN)6] (for a single-nanowire electrode) was used to determine the redox currents. The current–voltage (I–V) plots obtained for the SiNW devices by this method typically display a sigmoidal shape (Fig. 3c).

3.5. Lipid Bilayer Assembly and Protein Reconstitution on Carbon Nanotube Devices

1. For sodium pump reconstitution, prepare lipid vesicles made of DPhPC/DPhPE/PI/cholesterol as follows: Mix together aliquots of the appropriate lipid stock solutions (typically, 10  mg/ml in chloroform) to obtain a lipid mixture ­containing DPhPC/DPhPE/PI/cholesterol in the ratio of

Functional Integration of Membrane Proteins with Nanotube

547

61/14.4/2.6/22 (w/w/w/w) and a total amount of lipid of 2 mg. Dry the lipid mixture by gently flowing a stream of Ar for 1 h to remove the solvent. During this time, rotate the vial to form a thin lipid film on the walls. Rehydrate the dried lipid film in 2 ml of ATPase buffer containing 0.5 mg/ ml of C12E8. Vigorously vortex the resulting lipid/surfactant suspension and sonicate it in a water bath for 15 min. Keep the mixture cool in an ice bath. 2. To reconstitute the sodium pump, dissolve 1 mg of Na+,K+ATPase in 5 ml of cold ATPase buffer containing 0.5 mg/ml of C12E8 to produce a 0.2 mg/ml protein solution. Add 1 ml of this solution to 2  ml of the 1  mg/ml lipid/surfactant solution from Step 1 to achieve a 1:10 protein/lipid ratio by mass. Slowly remove the C12E8 surfactant from the resulting solution using Bio-Beads. Add 3 ml of Bio-Beads (120 mg/ml) to the protein/lipid/surfactant mixture and agitate with a stir bar for 12  h on ice. Allow the mixture to warm up to room temperature and remove the Bio-Beads. Repeat the incubation with fresh Bio-Beads for 1 h at room temperature to remove any remaining surfactant. To prepare control experiments in which the lipid bilayer does not contain Na+,K+-ATPase, use the same protocol but omit the protein addition step. 3. To form a lipid bilayer on the carbon nanotube devices, flow the lipid/protein vesicles through a PDMS microchannel placed over the device chip. To ensure complete and continuous coverage, allow the lipid bilayer to form over the course of 12 h at room temperature. Rinse away excess lipid/protein vesicles using ATPase buffer solution (without any surfactant). 3.6. Lipid Bilayer Assembly on Silicon Nanowire Devices

1. Add 40 ml of DOPC to a 4-ml glass vial with a Teflon-lined lid. (Alternatively, add a mixture containing a 1% molar ratio of NBDPE/DOPC if lipid bilayer imaging and/or fluidity measurements are to be carried out. When NBDPE is present in the lipid mixture, it is important that all the subsequent steps are performed with the glass vial covered with aluminum foil to avoid exposing the solution to light, which could result in photobleaching of the fluorescent probe.) 2. Evaporate the chloroform solvent by gently flowing pure N2 gas into the glass vial while rotating it continuously to form a thin film of lipid on the walls of the vial. 3. Once the solvent is no longer present, continue to flow N2 into the vial for an additional 30 min. 4. Hydrate the dried lipid film with 2 ml of K-PBS buffer, and then vortex the mixture in periods of 30 s until the lipids are completely dissolved. This procedure results in a solution containing a lipid concentration of 0.2 mg/ml.

548

Noy et al.

5. Using a probe tip sonicator, sonicate the lipid solution at 6 W for 30 s, followed by 3 W for 30 s to form unilamellar vesicles. Cool the vesicle solution to room temperature. 6. Deposit the vesicle solution onto a SiNW device chip (prepared as described in Subheading 3.3). Incubate the chip substrate with the vesicle solution for 24 h to allow the vesicles to fuse onto the SiNWs (see Note 14). 7. Rinse the chip substrate with copious amounts of K-PBS. Do not expose the coated substrate at any time to air bubbles to avoid destruction of the supported lipid bilayer structures. (Protect the substrate from light if the deposited bilayer is doped with fluorescent NBDPE probe molecules.) 3.7. Formation of a-Hemolysin Channels in Lipid Bilayers Assembled on Silicon Nanowire Devices

1. Add 1  ml of K-PBS solution to 0.5  mg of lyophilized a-hemolysin protein in a container. Caution: a-Hemolysin is a hemolytic toxin, so observe extreme care when handling! Gently agitate the solution by hand until all of the a-hemolysin protein is dissolved. Keep the protein stock solution at 4°C until use. 2. Take an aliquot of the a-hemolysin stock solution and dilute it to a final concentration of 100 mg/ml with K-PBS buffer. 3. Incubate a SiNW device chip precoated with a lipid bilayer (prepared as described in Subheading 3.6) with the 100 mg/ ml a-hemolysin solution for 18–24 h. 4. Rinse the SiNW chip substrate copiously with K-PBS buffer, while avoiding any exposure of the substrate to air bubbles.

3.8. Formation of Alamethicin Channels in Lipid Bilayers Assembled on Silicon Nanowire Devices

1. Dissolve alamethicin in K-PBS to obtain a final concentration of 5 mg/ml. Gently agitate the solution by hand until all of the alamethicin peptide is dissolved. Keep the stock solution at 4°C until use. 2. Dilute an aliquot of the alamethicin stock solution with K-PBS to give a final concentration of 30 mg/ml. 3. Incubate a SiNW device chip precoated with a lipid bilayer (prepared in Subheading 3.6) with the 30 mg/ml alamethicin solution for 30 min. 4. Rinse the SiNW chip substrate copiously with K-PBS buffer, taking care to avoid exposing the device to air bubbles.

3.9. Formation of Gramicidin A Channels in Lipid Bilayers Assembled on Silicon Nanowire Devices

1. Dissolve gramicidin A in ethanol to obtain a final concentration of 50 mg/ml. Gently agitate the solution by hand until all of the gramicidin A polypeptide is dissolved. 2. Add 40 ml of DOPC to a 4-ml glass vial with a Teflon-lined lid. Immediately add an aliquot of the 50  mg/ml gramicidin A solution (from Step 1) to the same glass vial to obtain a mixture containing DOPC/gramicidin A at a molar ratio of 100:1.

Functional Integration of Membrane Proteins with Nanotube

549

3. Evaporate the solvents by gently flowing pure N2 gas into the glass vial while rotating it continuously to form a thin film of lipid/polypeptide on the walls of the vial. 4. Once the solvents are no longer present, continue to flow N2 into the vial for an additional 30 min. 5. Hydrate the film of DOPC/gramicidin A with 2 ml of K-PBS buffer. 6. Incubate an unmodified SiNW device chip (prepared as described in Subheading 3.3) with the DOPC/gramicidin A solution for 24 h. 7. Rinse the SiNW chip substrate copiously with K-PBS buffer solution, taking care to avoid exposing the device to air bubbles.

4. Notes 1. Deionized water used for preparing phospholipid vesicles should be protected from atmospheric exposure or shaking; otherwise, it will form bubbles during the formation of supported phospholipid bilayers. 2. Other photoresists may also be used, but the tone of the resist (i.e., positive vs. negative) should be decided upon before drawing the mask, as the tones of the mask and photoresist should match. 3. A short exposure time to UV light followed by postbaking results in the formation of crosslinks that transform the otherwise positive photoresist AZ 5214-E to a negative one and results in image reversal. 4. Carbon nanotube growth – and thus the yield of successfully connected devices – is highly sensitive to the exact formulation of the Fe/Mo catalyst, such that even a 1-Å difference in the deposited layer thickness can have a dramatic effect. When performing nanotube growth experiments for the first time, it may be necessary to prepare an array of chip samples with gradually changing catalyst compositions to determine the formulation that yields optimal results. 5. A low chamber pressure and a fast rate are essential for the successful deposition of metallic Al; at higher pressures and/ or slower rates, the material is essentially deposited as aluminum oxide. 6. LOR-3A is insoluble in acetone. 7. The gas ratios can be varied to achieve different doping concentrations and, consequently, the electrical conductivity of

550

Noy et al.

the nanowires. For example, to synthesize heavily doped nanowires, 5 sccm of SiH4 (10% v/v in He) and 25 sccm of B2H6 (100 ppm in He) are flowed for 2 h. The resulting wires are typically larger in diameter, and indeed we have obtained 80–110 nm diameters using a gold nanoparticle catalyst size of 30 nm. Moreover, nanowires doped heavily with B2H6 gas during growth tend to exhibit amorphous silicon shells around a crystalline core upon TEM inspection. 8. The concentration of nanowires suspended in the solution can be varied by adjusting the total volume of ethanol in which the chips are sonicated and the sonication time. It is advisable not to sonicate the chips for more than a few seconds, as this tends to reduce the average length of the nanowires in suspension. Freshly prepared suspensions will typically yield a higher surface density of nanowires on chip substrates for device integration since nanowires suspended in an ethanolic solution tend to agglomerate after a few days. 9. The hard-bake is a critical process step, and so the hotplate temperature should be maintained within ±1°C of the desired temperature in order to produce consistent results. 10. Temperatures higher than 100°C tend to produce a CVD silicon nitride film with better overall qualities, but reflowing and thermal degradation of the photoresist can also occur under these conditions. We recommend that a low-stress dielectric film be deposited in order to avoid defects in the electrode passivation layer. 11. The thickness of the metal plus silicon nitride layers should be limited to approximately 2/3 of the thickness of the LOR-3A layer in order to facilitate the subsequent metal lift-off process. 12. We do not recommend using sonication during the lift-off process because it can result in the degradation of the contact between the silicon nanowires and the metal electrodes. 13. Longer anneal times or higher temperatures may lead to excess metal diffusion into the silicon nanowires, especially for Ni contacts. Metal diffusion, in turn, will lead to smaller lengths of the nanowires being sensitive to the surrounding solution. The forming gas anneal step also helps to passivate surface trap states and improves the overall electrical performance of the SiNW transistors. 14. The reference microelectrode must first be removed from the flow chamber prior to formation of the supported lipid bilayers, and then introduced again after the chamber has been rinsed. This procedure helps to avoid data misinterpretation due to the possible presence of lipid bilayers on the porous membrane of the electrode.

Functional Integration of Membrane Proteins with Nanotube

551

Acknowledgments A.N.’s research was supported by the U.S. Department of Energy, Office of Basic Energy Sciences, Division of Materials Sciences and Engineering; and by the University of California-Lawrence Livermore National Laboratory (UC-LLNL) Research Program. A.A., S-.H., and J.M. acknowledge support from the LLNLLawrence Scholar Program (LLNL-LSP). A.N. and N.M. acknowledge the use of the Molecular Foundry User Facility at Lawrence Berkeley National Lab (LBNL). Parts of this work were performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344. References 1. Corry, B. (2006) Understanding ion channel selectivity and gating and their role in cellular signalling. Mol. Biosyst. 2, 527–535. 2. Long, S.B., Campbell, E.B. and MacKinnon, R. (2005). Crystal Structure of a Mammalian Voltage-Dependent Shaker Family K + Channel. Science 309, 897–903. 3. Haupts, U., Tittor, J. and Oesterhelt, D. Closing in on bacteriorhodopsin: Progress in understanding the molecule. (1999) Annu. Rev. Biophys. Biomol. Struct. 28, 367–399. 4. Wu, L.Q. and Payne, G.F. (2004) Biofabrication: Using biological materials and biocatalysts to construct nanostructured assemblies. Trends in Biotechnol. 22, 593–599. 5. Zheng, G., Patolsky, F., Cui, Y., Wang, W.U., and Lieber, C.M. (2005) Multiplexed electrical detection of cancer markers with nanowire sensor arrays. Nat. Biotechnol. 23, 1294–1301. 6. Patolsky, F., Timko, B.P., Yu, G., Fang, Y., Greytak, A.B., Zheng, G., and Lieber, C.M. (2006) Detection, Stimulation, and Inhibition of Neuronal Signals with High-Density Nanowire Transistor Arrays. Science 313, 1100–1104. 7. Huang, Y., Duan, X., Cui, Y., Lauhon, L.J., Kim, K.H., and Lieber, C.M. (2001) Logic gates and computation from assembled nanowire building blocks. Science 294, 1313–1317. 8. Fromherz, P. and Stett, A. (1995) Siliconneuron junction: Capacitive stimulation of an individual neuron on a silicon chip. Phys. Rev. Lett. 75, 1670–1673.

9. Jenkner, M. and Fromherz, P. (1997) Bistability of membrane conductance in cell adhesion observed in a neuron transistor. Phys. Rev. Lett. 79, 4705–4708. 10. Kam, N.W.S., O’Connell, M., Wisdom, J.A., and Dai, H. (2005) Carbon nanotubes as multifunctional biological transporters and nearinfrared agents for selective cancer cell destruction. Proc. Natl. Acad. Sci USA 102, 11600–11605. 11. Kim, W., Ng, J.K., Kunitake, M.E., Conklin, B.R., and Yang P. (2007) Interfacing silicon nanowires with mammalian cells. J. Am. Chem. Soc. 129, 7228–7229. 12. Boxer, S.G. (2000) Molecular transport and organization in supported lipid membranes. Curr. Opin. Chem. Biol. 4, 704–709. 13. Alberts, B., Bray, D., Lewis, J., Raff, M., Roberts, K., and Watson, J. D. (1994) Molecular Biology of the Cell, 3rd edition. Garland Publishing, Inc., New York, NY. 14. Zhou, X., Moran-Mirabal J.M., Craighead, H.G., and McEuen, P.L. (2007) Supported lipid bilayer-carbon nanotube hybrids. Nat. Nanotechnnol. 2, 185–190. 15. Chen, X., Tam, U.C., Czlapinski, J.L., Lee, G.S., Rabuka, D., Zettl, A., Bertozzi, C.R. (2006) Interfacing carbon nanotubes with living cells. J. Am. Chem. Soc. 128, 6292–6293. 16. Martinez, J.A., Misra, N., Wang, Y., Stroeve, P., Grigoropoulos, C.P., and Noy, A (2009) Highly-efficient biocompatible single silicon nanowire electrodes with functional

552

Noy et al.

biological pore channels. Nano Lett. 9, 1121–1126. 17. Artyukhin, A.B., Stadermann, M., Friddle, R.W., Stroeve, P., Bakajin, O., Noy, A. (2006) Controlled electrostatic gating of carbon nanotube FET devices. Nano Lett. 6, 2080–2085. 18. Cui, Y., Lauhon, L.J., Gudiksen, M.S., Wang, J.F. and Lieber, C.M. (2001) Diameter-

controlled synthesis of single-crystal silicon nanowires. Appl. Phys Lett 78, 2214–2216. 19. Misra, N., Martinez, J.A., Huang, S.C., Wang, Y., Stroeve, P., Grigoropoulos, C.P., and Noy, A. (2009) Bioelectronic silicon nanowire devices using functional membrane proteins. Proc. Natl. Acad. Sci. USA 106, 13780–13784.

Chapter 34 Single-Step Conjugation of Antibodies to Quantum Dots for Labeling Cell Surface Receptors in Mammalian Cells Gopal Iyer, Jianmin Xu, and Shimon Weiss Abstract Labeling of cell surface receptors in living cells can be achieved using antibody-conjugated semiconductor quantum dots (QDs). The inherent photostable property of QDs can be exploited for understanding the arrangement and distribution of receptors in the plasma membrane. We describe herein methods that allow conjugation of antibodies to QDs in a single step without the formation of side products. This protocol can be adapted universally for any type of QD structure with a coating of free amino groups. Key words: Cell membrane labeling, Quantum dot, Peptides, Gel filtration, Epidermal growth factor receptor, Antibodies, Live cell imaging

1. Introduction Fluorescent proteins fused to gene products of interest can be observed in living cells for relatively short time scales (ms to ms) using advanced microscopy techniques before the common problems of fluorescence bleaching and degradation become apparent. To overcome this limitation, inorganic nanoparticle probes such as semiconductor quantum dots (QDs) – with their brighter optical properties, narrow full-width at half-maximum emission spectra, and reduced tendency to photobleach – have been applied in biomedical applications (1–3). However, since QDs are typically synthesized and stored in organic solvents and passivated by nonpolar ligands (4, 5), the exchange of hydrophobic to hydrophilic ligand capping groups is critical for biological applications. Consequently, members of our group at UCLA, as well as others, have developed efficient ligand exchange strategies that allow QDs to be capped by a hydrophilic layer (6–10),

Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_34, © Springer Science+Business Media, LLC 2011

553

554

Iyer, Xu, and Weiss

which results in either an amine or thiol surface coating. To effectively exploit the presence of free amino groups on the surface of the QDs for biofunctionalization, we have further extended a conjugation strategy previously reported for coupling oligonucleotides to peptides (11, 12). We describe herein a single-step method for functionalizing amine-capped QDs with an aromatic benzaldehyde linker that can be used for biorthogonal conjugation to hydrazinonicotinamide linker (HyNic)-modified anti-epidermal growth factor receptor (EGFR) antibodies. The reaction takes advantage of the presence of a N-hydroxysuccinimide ester (NHS-ester) present at one end of the linkers, which primarily reacts with free amino groups. The overall conjugation scheme ultimately produces a stable bis-aryl hydrazone bond between the QD and anti-EGFR antibody. The advantages of this approach over previous methods for the covalent coupling of biomolecules to QDs are as follows: (1) single-step purification following the reaction of the benzaldehyde linker with free amino groups on the QDs; (2) the benzaldehyde-modified QDs react specifically with HyNic-modified proteins even in the presence of other amine, carboxyl, or thiol groups; (3) the side product of the conjugation reaction is water; and (4) the conjugation reaction is carried out in aqueous buffers. The procedures outlined in this chapter can also be applied to other types of nanoparticle materials coated with surface amine groups. To demonstrate the selectivity and specificity of the anti-EGFR-QD conjugate synthesized in this example protocol, we chose to carry out a functional assay in living mammalian cells. In this work, the breast cancer cell line MCF-7 expressing human EGFR was labeled with anti-EGFR-conjugated QDs and imaged using standard confocal fluorescence microscopy; however, the anti-EGFR-QD conjugates may also be used in immunohistochemistry and flow cytometry applications, as well.

2. Materials 2.1. General Reagents, Buffers, and Other Materials

1. 1–2 nmol of CdSe/ZnS quantum dots (605 nm) in hexane (obtained commercially from Invitrogen, Carlsbad, CA or eBioscience, Inc., San Diego, CA; or synthesized in the laboratory according to previously published procedures, such as described in ref. (13)). 2. 1–2  nmol of amine (NH2)-functionalized QDs (eFluor® Nanocrystal (Amine), eBioscience). 3. Peptide (custom synthesized) for coating QDs: H2N-GlySer-Glu-Ser-Gly-Gly-Ser-Glu-Ser-Gly-Phe-Cys-Cys-PheCys-Cys-Phe-Cys-Cys-Phe-CONH2 (New England Peptide, Gardner, MA).

Single-Step Conjugation of Antibodies to Quantum Dots

555

4. S-4-Formylbenzamide (S-4FB) or PEG4/PFB (extended hydrophilic linker analog of S-4FB) (Solulink, San Diego, CA) (see Note 1). 5. S-Hydrazinonicotinamide (S-HyNic) (Solulink) (see Note 1). 6. Centrifugal filter concentrator devices (30  kDa MWCO) (VWR International Inc, Bridgeport, NJ). 7. Slide-A-Lyzer MINI dialysis unit (20 kDa MWCO) (Pierce, Rockford, IL). 8. Dimethyl sulfoxide (DMSO), anhydrous. 9. Pyridine, anhydrous. 10. Tetramethylammonium hydroxide (TMAOH), 25% (w/w) in methanol. 11. Microcentrifuge tubes (1.5 mL). 12. Illustra NAP™-25 desalting/buffer exchange columns (GE Healthcare, Piscataway, NJ). 13. 10× Conjugation buffer: 1 M 2-(N-morpholino)ethanesulfonic acid (MES) buffer, pH 6.0 (prepared in the laboratory or obtained from Solulink). 14. 10× Modification buffer: 1 M Phosphate buffer, 1.5 M NaCl, pH 7.4 (prepared in the laboratory or obtained commercially from Solulink). 15. Dialysis buffer: 100 mM Phosphate, 150 mM NaCl, pH 7.4 16. Protein molecular weight standards for gel filtration chromatography (No. 151-1901, Bio-Rad, Hercules, CA). 17. Superdex® 200 10/300 gel filtration column (GE Healthcare). 18. Gel filtration (HPLC/FPLC) buffer A: 100 mM phosphate, 150 mM NaCl, pH 7.4. 19. Gel filtration (HPCL/FPLC) buffer B: 100  mM borate, 200 mM NaCl, pH 8.3. 2.2. Antibodies and Buffers for Cell Labeling

1. Mouse anti-human EGFR antibody (BSA and gelatin free) (Santa Cruz Biotechnology, Santa Cruz, CA) in PBS, pH 7.4 (see Note 3). 2. 1× Hank’s Balanced Salt Solution (1× HBSS). 3. 1× HBSS plus 50 mM HEPES, pH 7.5. 4. Blocking buffer: 5% (w/v) Bovine serum albumin, 1% (w/v) Fetal bovine serum in 50  mM HEPES, pH 7.5, filter sterilized. 5. Wash buffer: 10 mM HEPES buffer, pH 7.5.

2.3. Cell Culture Materials

1. Dulbecco’s Modified Eagle’s medium (DMEM) with high glucose (No. 11995, Invitrogen), containing 1× penicillin/ streptomycin and 10% fetal bovine serum.

556

Iyer, Xu, and Weiss

2. MCF-7 human breast cancer epithelial cells (American Type Culture Collection, Manassas, VA). 3. Glass bottom culture dishes with 1.5-mm coverslips (No. P35G-1.5-14-C, MatTek Corp., Ashland, MA). 2.4. Analytical Instrumentation and Other Tools

1. Agilent 1100 HPLC system equipped with fluorescence and UV–visible detectors (see Note 2). 2. Handheld UV lamp (l = 360 nm). 3. UV–visible spectrophotometer. 4. Quartz cuvette (100-mL capacity, 10-mm path length, 2 × 2.5 mm window). 5. Fluorescence microscopy system equipped with either lampor laser-based illumination.

3. Methods 3.1. Synthesis and Purification of Peptide-Coated QDs Containing Free Amino Groups on the Surface 3.1.1. Synthesis of Peptide-Coated QDs

1. Precipitate CdSe/ZnS QDs from hexanes with a methanol: isopropanol (3:1, v/v) solution in a glass vial (see Note 4). 2. Transfer the solution containing the precipitated QDs to a microcentrifuge tube and centrifuge at maximum speed in a high-speed microcentrifuge for 2 min. Decant the supernatant into an inorganic waste container in the fume hood. 3. Wash the QD pellet with methanol. The volume of methanol added should be sufficient to completely cover the pellet. 4. Dissolve the QD pellet by adding 450 ml of anhydrous pyridine. 5. Transfer the solution of QDs in pyridine to a 1.5-ml microcentrifuge tube. 6. Measure the absorbance of the QD solution in a UV–visible spectrophotometer using a 100-ml quartz cuvette. The absorbance value should be ~0.2–0.3 OD at the first exciton peak of the QDs. For 605-nm emitting QDs, measure the absorbance at 595 nm. 7. In a separate 1.5-ml microcentrifuge tube, weigh out 4 mg of peptides. Next, dissolve the peptides in 50  ml of anhydrous DMSO. Ensure that the peptides are dissolved completely. 8. Pipette the entire solution (50 ml) containing dissolved peptides from step 7 into the microcentrifuge tube (from step 5) containing the QDs in pyridine (450 ml). 9. Add 11.6 ml TMAOH to the tube from step 8 and manually mix the solution rapidly for 10  s. The mixture will appear cloudy, and flocculent precipitates may also appear.

Single-Step Conjugation of Antibodies to Quantum Dots

557

10. Centrifuge the mixture at maximum speed for 2  min in a high-speed microcentrifuge. 11. Decant the supernatant into an inorganic waste container in the fume hood. A pellet should be visible at the bottom of the microcentrifuge tube. Ensure that the walls of the microcentrifuge tube are completely free of any solvent mixture. 12. Add 300 ml of anhydrous DMSO to the QD pellet. The pellet should dissolve slowly and homogenously. Do not attempt to vortex or tap violently. At this point, once the QD pellet dissolves completely, the peptide-exchange process is complete. 13. While the pellet is dissolving, equilibrate a NAP™-25 column with distilled or Milli-Q water four to five times following the manufacturer’s instructions. 14. Pipette the solution (300 ml) containing the QDs dissolved in DMSO into the NAP™-25 column and wait for the solution to enter the resin bed completely. 15. Elute the QDs with water while monitoring the band corresponding to the QDs with a handheld UV lamp (360  nm wavelength). 16. Collect 1 ml of eluent containing the peptide-coated QDs in a 1.5-ml microcentrifuge tube. At this stage, the QDs are completely solubilized in water and ready for use in conjugation applications (see Note 5). Alternatively, if desired, additional purification procedures may also be performed to remove residual peptides from the QD solution (see Subheading 3.1.2 below). 3.1.2. Purification of Peptide-Coated QDs

In order to remove residual peptides, the peptide-coated QDs synthesized in Subheading 3.1.1 are first dialyzed in aqueous buffer, and then further purified by gel filtration chromatography. 1. Dialyze the QD sample in dialysis buffer using a Slide-ALyzer MINI Dialysis Unit (20 kDa MWCO). Use the dialysis buffer at a total of at least 300 times the volume of the sample during the course of the dialysis procedure. Perform several changes of the dialysis buffer. 2. Concentrate the resulting dialysate from step 1 using a centrifugal filter concentrator device (30 kDa MWCO; 2,500 × g) to reduce the total sample volume to 300 ml. 3. Further purify the concentrated QD sample by performing gel filtration chromatography on a Superdex® 200 column (see Note 2). Load the entire sample amount (300 ml) onto the Superdex® 200 column, and elute the QDs in gel filtration buffer A (100 mM phosphate, 150 mM NaCl, pH 7.4) using a flow rate of 0.2–0.4  ml/min. This chromatography step eliminates any free peptides remaining in the QD solution.

558

Iyer, Xu, and Weiss

At this stage, the purified peptide-coated QDs with free amino groups (“chemical handles”) exposed on the surface can be conjugated to biomolecules using standard amine-based coupling chemistries. 3.2. Derivatization of Amine-Functionalized QDs with 4-FB Linker

The modification of amine-functionalized QDs with 4-FB linker molecules may be conveniently performed using either peptidecoated QDs synthesized in Subheading  3.1 above (see the modification procedures described below in Subheadings  3.2.1 and 3.2.2), or amine-functionalized QDs obtained from commercial suppliers (see the modification procedure described below in Subheading 3.2.3).

3.2.1. Reaction of S-4FB with Peptide-Coated QDs (Option 1)

1. Dissolve 10 mg of S-4FB in 500 ml of DMSO to give a stock solution of the linker. 2. In step 12 of Subheading  3.1.1 above, add a large molar excess of linker molecules in DMSO to the peptide-exchanged QD pellet (typically, the addition of 5–10  ml of the linker stock solution should be sufficient). 3. Incubate the S-4FB/QD mixture on a rotating shaker for 1–2 h. 4. Stop the derivatization reaction with 10 mM Tris–HCl (pH 7.5) or with any other equivalent amine-containing buffer (see Note 6). 5. Remove excess free linkers with a NAP™-25 column equilibrated in water by following the procedures described above in Subheading 3.1.1, steps 13–16. Optional: Further purify the 4-FB-modified QDs by performing the procedures described in Subheading 3.1.2 (use water for dialysis and elution during gel filtration), which will remove any residual free peptides. 6. Quantify the yield of 4-FB-modified QDs by measuring the absorbance using a UV–visible spectrophotometer (see Note 7).

3.2.2. Reaction of S-4FB with Peptide-Coated QDs (Option 2)

1. Prepare 1 nmol of purified peptide-coated QDs in 1× modification buffer (synthesized and purified as described in Subheading 3.1) and transfer into a 1.5-ml microcentrifuge tube. 2. Dissolve 10 mg of S-4FB in 500 ml of DMSO to give a stock solution of the linker. 3. Add a large molar excess of linker molecules in DMSO to the peptide-coated QD buffer solution (typically, the addition of 5–10 ml of the linker stock solution should be sufficient). 4. Incubate the S-4FB/QD reaction mixture on a rotating shaker for 30 min to 1 h.

Single-Step Conjugation of Antibodies to Quantum Dots

559

5. Stop the derivatization reaction with 10 mM Tris–HCl (pH 7.5) or with any other equivalent amine-containing buffer (see Note 6). 6. Remove excess free linker molecules by dialyzing the reaction solution against water with a Slide-A-Lyzer MINI dialysis unit (20 kDa MWCO). Perform dialysis using a total reservoir volume of at least 300 times the volume of the sample during the course of the dialysis procedure. Perform several changes of the dialysis buffer. 7. Quantify the yield of 4-FB-modified QDs by measuring the absorbance using a UV–visible spectrophotometer (see Note 7). 3.2.3. Reaction of S-4FB with Amine-Modified QDs

1. React S-4FB linkers with amine (NH2)-modified QDs (obtained from, e.g., eBioscience) by following the procedures described in Subheading 3.2.2, steps 1–5 above. 2. Remove excess free linker molecules by dialyzing the reaction solution against water with a Slide-A-Lyzer MINI dialysis unit (20 kDa MWCO). Perform dialysis using a total reservoir volume of at least 300 times the volume of the sample during the course of the dialysis procedure. Perform several changes of the dialysis buffer. 3. Quantify the yield of 4-FB-modified QDs by measuring the absorbance using a UV–visible spectrophotometer (see Note 7).

3.3. Derivatization of Antibodies with S-HyNic Linker

1. In a microcentrifuge tube, prepare a protein solution containing 1 mg/ml unmodified anti-EGFR antibodies in 1× modification buffer (see Note 3). 2. In a separate microcentrifuge tube, prepare a linker stock solution containing 20 mg/ml S-HyNic in anhydrous DMSO. 3. Add a sufficient quantity of the S-HyNic stock solution to the protein solution to produce an eight- to tenfold molar excess of S-HyNic linker molecules over the antibody concentration. 4. Incubate the reaction mixture for 1 h at room temperature or for 16–24 h at 4°C on a rotating shaker, depending on the type of antibody to be derivatized. 5. During the reaction incubation period, equilibrate a 2-ml NAP™-25 desalting column with 1× conjugation buffer. 6. Stop the derivatization reaction with 10  mM Tris–HCl (pH 7.5) or with any other equivalent amine-containing buffer (see Note 6). 7. Desalt the HyNic-modified anti-EGFR antibodies in 1× conjugation buffer using the equilibrated NAP™-25 column prepared in step 5 above to remove excess free linker molecules.

560

Iyer, Xu, and Weiss

3.4. Conjugation of 4-FB-Modified QDs to HyNic-Modified Antibodies

1. Add a sufficient volume of 10× conjugation buffer to the stock solution of 4-FB-modified QDs in water to give a final solution containing 1× conjugation buffer. 2. Add a sufficient volume of HyNic-modified anti-EGFR antibodies in 1× conjugation buffer to the solution of 4FB-modified QDs in 1× conjugation buffer to give a 4:1 molar ratio of antibodies:QDs (see Note 8). 3. Incubate the reaction mixture on a rotating shaker overnight at 4°C. 4. Equilibrate a Superdex® 200 column with gel filtration buffer B (100 mM Borate, 200 mM NaCl, pH 8.3). 5. Calibrate the Superdex® 200 gel filtration column by running the protein molecular weight standards first, followed by the unconjugated QDs. Adjust the flow rate to 0.3 ml/min and record the elution times for each run (see Note 9). 6. Load the entire reaction mixture onto the Superdex® 200 gel filtration column and separate the anti-EGFR antibodyconjugated QDs from the unconjugated QDs and free antiEGFR antibodies. (Adjust the column flow rate to 0.3 ml/min.) 7. Monitor the elution profile by absorbance measurements at 280 nm wavelength. The first peak should correspond to the antibody-conjugated QDs, the second peak corresponds to unconjugated QDs, and the third peak corresponds to free antibodies (see Note 9). 8. Store the eluted anti-EGFR antibody-conjugated QDs (antiEGFR-QDs) at 4°C. Under these storage conditions, the antibody–QD conjugates are typically stable for ~6–8 months.

3.5. Preparation of Cell Cultures for Labeling Live Cells with QD-Conjugated Antibodies

1. Remove a single vial containing a frozen stock of adherent MCF-7 cells from vapor phase liquid nitrogen storage and thaw the cells in a 37°C water bath. 2. Add 10 ml of prewarmed DMEM medium and centrifuge the cells at 800 × g for 5 min. 3. Resuspend the pellet in 1 ml of DMEM medium by gently pipetting up and down a few times. Prepare a solution containing a sufficient concentration of cells to give ~30–40% confluence after 12 h. 4. Incubate the cells in glass bottom dishes. After 12  h, add fresh DMEM medium and ensure that the adherent cells have completely flattened out.

3.6. Staining Live Cells with QD-Conjugated Antibodies

1. Replace the DMEM media in the glass bottom dishes containing adhered cells with 1× HBSS and aspirate twice. 2. Incubate the cells with blocking buffer at 37°C for 30–60 min. Monitor the cells periodically under phase contrast microscopy to ensure that they maintain a normal appearance.

Single-Step Conjugation of Antibodies to Quantum Dots

561

Fig. 1. Quantum dot labeling of live MCF-7 breast cancer epithelial cells. (a) Fluorescent image demonstrating the specificity of cell surface targeting of EGFR receptors with anti-EGFR antibodies conjugated to CdSe/ZnS QDs. (b) Differential interference contrast image of the same area shown in (a).

3. Aspirate once and wash the cells with 1 ml of 10 mM HEPES buffer, pH 7.5. 4. Incubate the cells for 10 min in 1× HBSS (or serum-containing medium) containing 1–2 nM anti-EGFR-QDs (see Note 10). 5. Prior to imaging (see Subheading  3.7 below), remove unbound antibody–QD conjugates by repeated aspiration with 1× HBSS plus 50 mM HEPES, pH 7.5. At this stage, one may proceed to observe the distinctive membrane labeling of EGFR cell surface receptors in the live breast cancer cell line MCF-7 by using fluorescence microscopy (see Subheading 3.7 below and Fig. 1). 3.7. Imaging Live Cells Labeled with QD-Conjugated Antibodies

Imaging of live cells labeled with antibody–QD conjugates can be performed using fluorescence microscopy systems that employ either lamp- or laser-based illumination. For live cell imaging over long time periods, it is important to maintain the cells at 37°C under a 5% CO2 atmosphere. Since QDs have a broadband excitation, it is advantageous to excite the samples at wavelengths close to 405  nm to minimize autofluorescence; however, a tradeoff between potentially damaging cell samples with exposure to UV light and minimizing autofluorescence should be taken into consideration when choosing the most appropriate filter set to use.

4. Notes 1. After dissolving into solution, store the linkers either in a glovebox or under nitrogen since the NHS-ester group of the linkers will become hydrolyzed upon slight exposure to air. 2. For the gel filtration chromatography procedure described in Subheading 3.1.2, step 3, the Superdex® 200 column may be used with either an HPLC or an FPLC system. It is important

562

Iyer, Xu, and Weiss

to maintain low flow rates of ~0.2–0.4 ml/min for optimal resolution during separation. Moreover, calibration of the gel filtration column with molecular weight markers after each run is important to perform, as this will help rule out if aggregates of QDs are stuck or have interacted nonspecifically with the column resin. 3. Full-length or F(ab)2-fragment antibodies may also be used for the conjugation procedures. It is important to ensure that the antibody stock solution is free of any preservatives since this may reduce the final yield of the conjugation reactions. 4. QDs from hexane can be purchased commercially from various suppliers. When using TOPO-coated QDs, precipitation can be achieved with methanol. With QDs coated with oleic acid or long-chain fatty acids, repeated fluxing with chilled acetone can be used to eliminate the inorganic coating. 5. Peptide-coated QDs obtained at this stage are ready for use with most of the standard amine-based conjugation chemistry procedures. 6. Termination of NHS-ester reactions with an amine buffer should be carried out using a 1 M stock solution. 7. It is important to have at least 1 nmol of peptide-coated or amine-coated QDs to use for any successful conjugation procedure. 8. The optimum ratio of antibodies to QDs to use should be determined empirically. Using a four- to tenfold molar excess of antibodies to QDs usually provides an adequate level of separation with respect to conjugated versus unconjugated QDs during gel filtration. 9. To increase the column life, it is important to check the elution profile of the molecular weight standards after each run. Different QD preparations obtained from various commercial vendors can have batch-to-batch variations, and this can affect the gel permeation characteristics of the column since QDs may interact with the column resin in a nonspecific manner. 10. Cell labeling with anti-EGFR-QDs should be performed using concentrations in the range of 1–10 nM. The important step here is to incubate the conjugate in the center of the (glass bottom) dish, which can hold approximately 100 ml of liquid, as opposed to using the entire surface area. Alternatively, one may also test the performance of the antibody-conjugated QDs via titration experiments using flow cytometry.

Single-Step Conjugation of Antibodies to Quantum Dots

563

Acknowledgments These protocols were developed in part by support from grants NIH/NIBIB BRP 5-R01-EB000312 and NIH 1-R01GM086197-01. Ensemble confocal imaging was performed at the UCLA/CNSI Advanced Light Microscopy Shared Facility. References 1. Michalet, X., Pinaud, F. F., Bentolila, L. A., Tsay, J. M., Doose, S., Li, J. J., Sundaresan, G., Wu, A. M., Gambhir, S. S., Weiss, S. (2005) Quantum dots for live cells, in  vivo imaging, and diagnostics. Science 307, 538–544. 2. Alivisatos, P. (2004) The use of nanocrystals in biological detection. Nat. Biotechnol. 22, 47–52. 3. Kim, S., Lim, Y. T., Soltesz, E. G., De Grand, A. M., Lee, J., Nakayama, A., Parker, J. A., Mihaljevic, T., Laurence, R. G., Dor, D. M., Cohn, L. H., Bawendi, M. G., Frangioni, J. V. (2004) Near-infrared fluorescent type II quantum dots for sentinel lymph node mapping. Nat. Biotechnol. 22, 93–97 4. Li, J. J., Wang, Y. A., Guo, W., Keay, J. C., Mishima, T. D., Johnson, M. B., Peng, X. (2003) Large-scale synthesis of nearly monodisperse CdSe/CdS core/shell nanocrystals using air-stable reagents via successive ion layer adsorption and reaction. J. Am. Chem. Soc. 125, 12567–12575. 5. Talapin, D. V., Rogach, A. L., Kornowski, A., Haase, M., Weller, H. (2001) Highly luminescent monodisperse CdSe and CdSe/ZnS nanocrystals synthesized in a hexadecylaminetrioctylphosphine oxide-trioctylphospine mixture. Nano Letters 1, 207–211. 6. Pinaud, F., King, D., Moore, H. P, Weiss, S. (2004) Bioactivation and cell targeting of semiconductor CdSe/ZnS nanocrystals with phytochelatin-related peptides. J. Am. Chem. Soc. 126, 6115–6123.

7. Iyer, G., Pinaud, F., Tsay, J., Weiss, S. (2007) Solubilization of quantum dots with a recombinant peptide from Escherichia coli. Small 3, 793–798. 8. Kim, S., Bawendi, M. G. (2003) Oligomeric ligands for luminescent and stable nanocrystal quantum dots. J. Am. Chem. Soc. 125, 14652–14653. 9. Liu, W., Howarth, M., Greytak, A. B., Zheng, Y., Nocera, D. G., Ting, A. Y., Bawendi, M. G. (2008) Compact biocompatible quantum dots functionalized for cellular imaging. J. Am. Chem. Soc. 130, 1274–1284. 10. Uyeda, H. T., Medintz, I. L., Jaiswal, J. K., Simon, S. M., Mattoussi, H. (2005) Synthesis of compact multidentate ligands to prepare stable hydrophilic quantum dot fluorophores. J. Am. Chem. Soc. 127, 3870–3878. 11. Dirksen, A., Dirksen, S., Hackeng, T. M., Dawson, P. E. (2006) Nucleophilic catalysis of hydrazone formation and transimination: implications for dynamic covalent chemistry. J. Am. Chem. Soc. 128, 15602–15603. 12. Dirksen, A., Dawson, P. E. (2008). Rapid oxime and hydrazone ligations with aromatic aldehydes for biomolecular labeling. Bioconjug. Chem. 19, 2543–2548. 13. Dabbousi, B.O., Rodriguez-Viejo, J., Mikulec, F.V., Heine, J.R., Mattoussi, H., Ober, R., Jensen, K.F., and Bawendi, M.G. (1997) (CdSe)ZnS core-shell quantum dots: Synthesis and characterization of a size series of highly luminescent nanocrystallites. J. Phys. Chem. B 101, 9463–9475.

Chapter 35 A Practical Strategy for Constructing Nanodrugs Using Carbon Nanotubes as Carriers Wei Wu and Xiqun Jiang Abstract Carbon nanotubes, acting as nanocarriers, can be combined with drug molecules through various chemical or physical routes in which hydrophilic modifications of the nanotubes are required. Such types of hydrophilic modifications typically involve addition reactions at the nanotube sidewalls, the reactions of nanotube-bound carboxylic groups, and/or coating of the nanotubes with amphiphilic molecules. In this chapter, we introduce detailed approaches for covalently linking drug compounds to multiwalled carbon nanotubes, as well as labeling the synthesized drug-bearing carbon nanotube conjugates with fluorescent or radioactive molecules. Key words: Multiwalled carbon nanotubes, Anticancer drug delivery, 10-Hydroxycamptothecin, Fluorescein isothiocyanate, Technetium-99m (99mTc), Single-photon emission computed tomography imaging

1. Introduction The utilization of biocompatible nanomaterials in drug delivery systems has brought about new hope in the development of novel, effective treatments for cancer. The rational design, modification, and incorporation of these nanomaterials potentially allows the purposeful manipulation of the pharmacological profiles of clinical drug compounds, making it possible to maximize their therapeutic efficacy while minimizing unwanted side effects (1). Among the diversity of nanomaterials currently being explored for biomedical applications, carbon nanotubes have been considered as one of the promising types of drug

Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_35, © Springer Science+Business Media, LLC 2011

565

566

Wu and Jiang

­ anocarriers due to their quasi-one-dimensional nanostructure, n full ­carbon composition, unique optical and electronic properties, ultrahigh surface area, remarkable cell membrane penetrability, and facile functionalization by a range of methods (2–13). On the other hand, there are also several factors that could significantly hamper the application of carbon nanotubes in the biomedical field. For example, while it has been generally accepted that well-purified, hydrophilic carbon nanotubes exhibit good biocompatibility (2), the presence of metal contaminants derived from catalysts used in carbon nanotube synthesis reactions and the tendency of carbon nanotubes to undergo aggregation in aqueous media could potentially induce significant toxicity (2, 3, 14, 15). Furthermore, the hydrophobic surface properties of carbon nanotubes makes them very susceptible to opsonizationinduced reticuloendothelial system uptake processes, which can significantly reduce their retention time in the bloodstream and consequently the chance of reaching pathological target tissues (2, 3, 16). Thus, it is important to remove metal contaminants and improve the hydrophilicity of carbon nanotubes in order to promote their successful application in biomedicine. For this ­purpose, various chemical or physical approaches for preparing functionalized carbon nanotubes have been developed over the past two decades. For example, metal contaminants are commonly eliminated by treatment with strong oxidizing agents, such as HNO3, HNO3/H2SO4, KMnO4/H2SO4, K2Cr2O7/H2SO4, oxygen gas, or OsO4, etc. (17), and the conversion of carbon nanotubes from a hydrophobic state to a hydrophilic one is commonly achieved by covalent/noncovalent chemical functionalization with hydrophilic molecules via reactions with nanotube-bound carboxylic acid groups, other types of sidewall reactions of carbon nanotubes, or coating with amphiphilic molecules bearing functional groups (2). In this chapter, we introduce a practical strategy for synthesizing hydrophilic drug-bearing multiwalled carbon nanotube (MWNT) conjugates labeled with fluorescent or radioactive molecules. In this approach, the amidation reaction of carbon nanotube-bound carboxylic acid groups is used as a starting point, due to its well-demonstrated effectiveness in chemically modifying carbon nanotube structures. A hydrophilic spacer comprised of diaminotriethylene glycol is also employed to improve the hydrophilicity of the nanotube conjugate structures. And finally, to enable the determination of their location within living tissues/whole organisms, the drug-bearing MWNT conjugates are labeled with either fluorescein isothiocyanate (FITC) or the radioactive nuclide of technetium-99m (99mTc) chelated by diethylenetriaminepentaacetic acid (DTPA).

A Practical Strategy for Constructing Nanodrugs

567

2. Materials 1. MWNTs with a diameter range of 10–30  nm and a purity level of ³95% can be obtained from commercial vendors. Alternatively, MWNTs can be synthesized in the laboratory by standard chemical vapor deposition techniques. Store the MWNTs at room temperature. 2. 10-Hydroxycamptothecin (HCPT) is a yellow powder and is light-sensitive. Store at 4°C. 3. 4-Dimethylaminopyridine (DMAP). 4. Succinic anhydride. 5. Chloroform (CHCl3). 6. Methanol (CH3OH). 7. 2,2¢-(Ethylenedioxy)diethylamine is a colorless liquid. Store at 4°C. 8. Di-tert-butyl dicarbonate. Store at −20°C. 9. Dichloromethane (DCM). 10. Diethyl ether. 11. Triethylamine (Et3N). 12. N-(3-Dimethylaminopropyl)-N ¢-ethylcarbodiimide hydrochloride (EDC·HCl) is a white powder. Store at −20°C. 13. N-Hydroxysuccinimide (NHS) is a colorless crystal. Store at 4°C. 14. Thionyl chloride. Distill before use and store in a desiccator. 15. Fluorescein isothiocyanate. 16. Technetium-99m (99mTc). 17. Diethylenetriaminepentaacetic acid. 18. Water bath sonicator (100 W). 19. Polycarbonate membrane filter, 200-nm pore size (Whatman). 20. Nylon-66 membrane filter, 200-nm pore size (Whatman). 21. Kaiser test reagent A: Mix 0.5 ml of 0.065% (w/v) KCN in water with 24.5 ml of dry pyridine and 2.5 ml of 400% (w/v) phenol in absolute ethanol. 22. Kaiser test reagent B: Dissolve 2.5 g of ninhydrin in 50 ml of absolute ethanol. Store the solution in the dark under nitrogen (N2). 23. Chromatography paper (Whatman No. 1). 24. Nuclear magnetic resonance (NMR) spectrometer. 25. Electrospray ionization mass spectrometer (ESI-MS). 26. UV–visible spectrophotometer.

568

Wu and Jiang

3. Methods As-prepared MWNTs are highly hydrophobic, typically ­contaminated with metal catalyst particles, and do not bear any significant quantity of active functional groups. It is thus pivotal to overcome these shortcomings via chemical or physical methods before carbon nanotubes can be utilized as drug carriers. Oxidation treatment has been frequently demonstrated to be effective in removing metal contaminants and simultaneously yielding carboxylic acid groups at defect sites along the sidewalls and at the ends of carbon nanotubes. The carboxylic groups thus produced can be employed to covalently link hydrophilic molecules, drugs, or other functional molecules. Moreover, oxidation treatment can also result in the physical cutting of nanotubes so that they become significantly shorter. In this section, we present a detailed strategy for the synthesis of water-soluble MWNT-drug conjugates, in which the ­antitumor agent HCPT is selected as a model drug compound, and a hydrophilic spacer comprised of diaminotriethylene glycol is introduced between the nanotube and the drug molecule. The overall synthetic route is summarized in Fig. 1. The carboxylic acid groups produced by the initial oxidative treatment of the carbon nanotubes are used as the starting point for the entire chemical functionalization procedures. To improve the hydrophilicity of the carbon nanotubes, it is important to increase the amount of nanotube-bound carboxylic acid groups so that a sufficient quantity of hydrophilic and functional molecules can be linked. As discussed in Subheading 3.3, the most suitable reaction conditions leading to the greatest amount of MWNT-bound carboxylic acid groups can be determined by studying the specific functional relationship between the duration of the oxidation treatment and the quantity of functional groups created. To covalently immobilize HCPT to MWNTs, a succinatebased HCPT ester derivative (d-HCPT) is initially synthesized (Fig.  1). The d-HCPT is then linked to MWNTs via two synthetic steps: First, amino groups are introduced to the surface of the MWNTs through an amidation reaction between the nanotube-bound carboxylic acid groups and tert-butyloxycarbonyl (Boc)-monoprotected diaminotriethylene glycol, followed by cleavage of the Boc group. Second, d-HCPT activated previously with EDC·HCl and NHS is attached to the surface of the aminomodified MWNTs 2 (once again via amidation), affording the d-HCPT–MWNT conjugate 3. Finally, the remnant amino groups present on the carbon nanotubes are used further to link fluorescent or radioactive labels (Fig. 1).

A Practical Strategy for Constructing Nanodrugs HO

N

N

O

O O

O

HO

succinic anhydride

O

N

4-dimethylaminopyridine, dry DMF

N

569

O O

OHO

OHO d-HCPT

a,b,c

O

O

O

NH3+

d

MWNTs 1

MWNTs 2 O

O

NH

OH

O

O

N

O

O

O

O

N

O

O

NH

O

O

N H

OH

O

NH

O

O

O

OHO

N N

NH

O O

e

HN S HN N H

O

HN

O

O

O

O

H2N

MWNTs 3

HOOC O

O

O

f,g

OH

MWNTs 4 O O O

NH

O

O

O

OHO

N

O

N O

NH

HN

O

O O

O

N O O

ON N 99mTc O

O

O

H N

O

O MWNTs 5

Fig. 1. Synthesis route for the preparation of fluorescent- and radiolabeled MWNT–HCPT conjugates. (a) Thionyl chloride, reflux; (b) Boc-NH(CH2CH2O)2–CH2CH2NH2, triethylamine, anhydrous THF, reflux; (c) 4  M HCl in dioxane; (d) d-HCPT, EDC·HCl, NHS, triethylamine, anhydrous DMF; (e) FITC, anhydrous DMF; (f) DTPA dianhydride, triethylamine, anhydrous DMSO; and (g) stannous chloride, 99mTcO4−. Reproduced with permission from (20) © 2009 American Chemical Society.

570

Wu and Jiang

3.1. Synthesis of Succinate-Based HCPT Ester Derivative (d-HCPT) (Fig. 2)

The succinate-based HCPT ester derivative (d-HCPT) is synthesized via an esterification reaction between succinic anhydride and the 10-hydroxy group of HCPT (see Note 1). 1. 104.3 mg (0.29 mmol) of HCPT, 35.0 mg (0.29 mmol) of DMAP and 31.0 mg (0.31 mmol) of succinic anhydride are dissolved in 3.6 ml of anhydrous DMF. The resultant mixture is stirred in darkness overnight at room temperature, and then an additional 7.8  mg (0.078  mmol) of succinic anhydride is added. The reaction mixture is stirred under the same conditions for another 5 h. 2. After the 5-h incubation period, an aqueous methanol solution (20% v/v, 0.5  ml) is added, and the stirring process is continued for additional 30  min to hydrolyze the excess anhydride. 3. The solvents are evaporated under vacuum below 40°C to obtain an oily residue. 4. To the oily residue is then added a solution of aqueous methanol (50% v/v, 5  ml). After a brief sonication treatment at room temperature, the obtained yellow slurry is allowed to stand for 5 h at 4°C. 5. The precipitated material is collected by centrifugation and dried under vacuum. 6. The crude product is purified by silica gel chromatography (eluent: CHCl3/CH3OH from 93/7 to 90/10, v/v) and recrystallized from methanol to give the pure d-HCPT product as a pale yellow solid (69.8 mg, 52%). Characterization of d-HCPT by thin-layer chromatography (TLC): Rf = 0.25 (eluent: CHCl3/CH3OH 90/10 (v/v)). 1H NMR (500 MHz, DMSO-d6) d: 0.91 (t, 3H), 2.14 (t, 2H), 2.47 (t, 2H), 2.69–2.82 (m, 2H), 5.21 (s, 2H), 5.46 (s, 2H), 7.02 (s, 1H), 7.27 (s, 1H), 7.42 (d, 1H), 8.02 (d, 1H), 8.43 (s, 1H), 10.57 (br s, 1H). 13C NMR (125 MHz, DMSO-d6) d: 7.51, 28.50, 28.63, 30.38, 50.05, 66.27, 75.88, 94.17, 108.77, 117.97, 123.04, 129.20, 129.68, 129.81, 130.58, 143.16, 145.29, 146.31, 149.18, 156.54, 156.69, 167.19, 171.23, 173.02. ESI-MS (m/z): Calcd. 463.11; found: 463.08 for (M − H)−.

HO N

N

O

O succinic anhydride O

DMAP, dry DMF

HO

O O

N

OHO

N

O O OH O

d-HCPT

Fig. 2. Synthesis of succinate-based HCPT ester derivative (d-HCPT).

A Practical Strategy for Constructing Nanodrugs

571

O O H2N

O

O

NH2

O O O 1,4-dioxane

H2N

O

O

O N O H

Fig. 3. Synthesis of Boc group-monoprotected ­diaminotriethylene glycol.

3.2. Synthesis of Boc-Monoprotected Diaminotriethylene Glycol (Fig. 3)

Boc group-monoprotected diaminotriethylene glycol is used to synthesize amino group bearing MWNTs 2 in order to avoid bridging of the nanotubes by the unprotected form of the diamine compound (see Note 2). 1. To a stirred solution of 2,2¢-(ethylenedioxy)bis(ethylamine) (10 g, 68 mmol) in 50 ml of 1,4-dioxane is added dropwise a solution of di-tert-butyl dicarbonate (1.47  g, 6.8  mmol) in 30 ml of 1,4-dioxane over 3 h. The resulting mixture is stirred at room temperature overnight. 2. After the evaporation of the solvent under reduced pressure, the residue is dissolved in 100 ml of DCM, washed with water (3 × 30 ml), and then dried with anhydrous Na2SO4. 3. Thereafter, the crude product is purified by silica gel flash chromatography with an eluent of DCM/methanol/triethylamine (Et3N) (80/20/1, v/v/v). Characterization of Boc-monoprotected diaminotriethylene glycol by 1H NMR (300  MHz, CDCl3) d: 1.39 (s, 9H), 2.0 (s, 2H), 2.84 (t, 2H), 3.26 (m, 2H), 3.45–3.51 (m, 4H), 3.57 (s, 4H), 5.20 (br, 1H).

3.3. Synthesis of Amine-Bearing MWNTs

3.3.1. Oxidation of MWNTs (Fig. 4)

To improve the hydrophilicity and the drug loading capacity of  the MWNTs, it is important to fully increase the amount of nanotube-bound carboxylic acid groups, which can be achieved by systematically varying the oxidation treatment duration and selecting the optimal reaction conditions accordingly. 1. To a clean 250-ml round-bottom flask are added 2 g of pristine MWNTs and 100  ml of a mixture of concentrated H2SO4/HNO3 (3/1, v/v). 2. The resulting acidic mixture from step 1 is sonicated for 60 h by using an ultrasonic water bath with a power of 100 W (see Note 3), while maintaining the temperature of the water bath below 60°C. 3. Aliquots (10 ml) are removed from the acidic reaction mixture during the sonication treatment (step 2) at predetermined times with an interval of 12 h. In doing so, five distinct samples of MWNTs – subjected to 12, 24, 36, 48, or 60 h of oxidation treatment – are obtained. Each of the oxidized MWNT samples is immediately diluted with ten volumes of deionized water as soon as it is removed from the acidic reaction mixture.

572

Wu and Jiang

H2SO4/HNO3, (3/1, v/v) sonication

O

O

thionyl chloride OH

Cl

reflux

HN O

1

dry THF,Et3N, reflux H2N

O N H

O

O

NH3+

MWNTs 7

4 M HCl in dioxane

O N H

O

O O

O

O

H N O O

MWNTs 6

Fig. 4. Oxidization of MWNTs and reaction of nanotube-bound carboxylic acid groups with Boc group-monoprotected diaminotriethylene glycol.

Fig.  5. SEM image of MWNTs after oxidation treatment for 36  h. Reproduced with p­ ermission from (20) © 2009 American Chemical Society.

4. The oxidized MWNT samples collected from step 3 are ­centrifuged to remove excess acid and water. 5. The black solid pellet of MWNTs obtained after centrifugation (step 4) is resuspended in ~3 ml of deionized water, filtered through a polycarbonate filter (200 nm pore size), and then rinsed thoroughly with deionized water until the pH is ~6. 6. The oxidized MWNT product is obtained as a black solid after drying under vacuum at 50°C. After completing the above procedures, we have observed that the length of the MWNTs oxidatively treated for 36 h is in the range of 20–700 nm, with a mean length of ~160 nm as statistically estimated by scanning electron microscopy (SEM) analyses (Fig. 5).

A Practical Strategy for Constructing Nanodrugs 3.3.2. Measurement of the Amount of Nanotube-Bound Carboxylic Groups Generated in Oxidized MWNTs

573

The carboxylic acid group content in the samples of oxidized MWNTs obtained from Subheading 3.3.1 is assessed by using a quantitative Kaiser test (18) (see Note 4). In order to perform this analysis, the nanotube-bound carboxylic acid groups are first reacted with Boc-monoprotected diaminotriethylene glycol, which is then followed by the cleavage of the Boc groups to generate amine-bearing MWNTs 7 (Fig. 4). The Kaiser test is a colorimetric assay that reveals quantitatively the presence of primary amino groups and, therefore, can be used to determine the number of carboxylic acid functional groups originally attached to the oxidized MWNTs. In the procedures below, nanotube samples derived from different batches of oxidized MWNTs are treated in parallel in order to minimize any potential variability in the amidation reaction conditions. 1. 50 mg of oxidized MWNTs (obtained from Subheading 3.3.1) are dispersed in 10 ml of thionyl chloride by sonication and refluxed for 24 h. 2. After evaporation of the excess thionyl chloride under vacuum, the obtained black solid is suspended in a solution of Boc-NH(CH2CH2O)2–CH2CH2NH2 (1, 250 mg) in 5 ml of anhydrous THF containing 50 ml of triethylamine (Et3N) and heated under reflux and argon for 48 h (see Note 5). 3. After removal of the solvents by evaporation under vacuum, the crude product is purified by washing thoroughly with methanol and diethyl ether. 4. The removal of excess 1 is monitored by TLC (eluent: DCM/ methanol/Et3N, 80/20/1 (v/v/v)). A 0.2% ninhydrin ethanol solution is used to visualize the TLC plate after heating at 100°C. 5. After drying at 50°C under vacuum, the obtained MWNTs 6 (Fig. 4) are suspended in 6 ml of 4 M HCl in 1,4-dioxane and stirred at room temperature for 5 h to cleave the Boc protecting group at the chain end. 6. After the 5-h incubation period, the solvent is evaporated from the reaction mixture. 7. The resulting amine-modified MWNT product 7 (Fig. 4) is then washed with diethyl ether several times and dried under vacuum. 8. The amount (loading) of amino groups per gram of MWNTs 7 is measured by performing a quantitative Kaiser test. Approximately 0.1–0.3  mg of MWNTs 7 are placed inside a glass test tube, followed by the addition of 100 mL of Kaiser test Reagent A and 25 mL of Kaiser test Reagent B. A control (blank) sample is prepared by adding the same amounts of reagents A and B to another test tube in the absence of MWNTs 7.

574

Wu and Jiang

After the solutions are mixed well, both test tubes are heated to 100°C for 10 min and then placed in a cold water bath. 1 ml of 60% ethanol in water is added to each test tube and mixed thoroughly. The solution in each test tube is filtered through a Pasteur pipette containing a tight plug of glass wool. The glass wool is rinsed twice with 0.2 ml of 0.5 M tetraethylammonium chloride in DCM, and the recovered filtrate solution is diluted to a total volume of 2.00 ml with 60% ethanol. The absorbance of the filtrate solution is measured against the reagent blank at 570 nm wavelength. The concentration of  free amines in the MWNTs 7 can be calculated by using c = A/(e × l), where c is the concentration (M), A is the absorbance, e is the effective extinction coefficient (1.5 × 104 M−1 cm−1) (18), and l is the optical path length of the cuvette (1 cm). The relationship between the calculated loading of primary amino groups on the MWNTs 7 and the oxidation treatment duration is shown in Fig. 6. Based on this relationship, it can be inferred that the amount of carboxylic acid groups initially present on the carbon nanotubes (i.e., before derivatization with compound 1) generally increases with the length of the oxidation treatment, from 0.41  mmol/g for a 12-h treatment to 0.63 mmol/g for a 36-h treatment; when the oxidation duration

Fig. 6. Amount of functional groups loaded onto MWNTs after various oxidation treatment times. The loading values were obtained by the quantitative Kaiser test after the introduction of primary amine groups via reaction of the nanotube-bound carboxylic acid groups with Boc group-monoprotected diaminotriethylene glycol. Reproduced with permission from (20) © 2009 American Chemical Society.

575

A Practical Strategy for Constructing Nanodrugs

exceeds 36 h, however, the loading tends to remain approximately constant (see Note 6). Accordingly, MWNTs oxidized for 36 h were selected for the subsequent synthesis procedures described below in Subheadings 3.4–3.6. 3.4. Functionalization of Amino-Modified MWNTs with HCPT (Fig. 7)

HCPT is linked to amine-modified MWNTs 2 (Fig.  7) via an amidation reaction between activated d-HCPT molecules and nanotube-bound free amino groups to generate MWNT–HCPT conjugates 3 (Fig. 7). 1. d-HCPT (87.6 mg, 0.19 mmol) in 5 ml of anhydrous DMF containing 205.6 ml (1.43 mmol) of triethylamine is activated with EDC·HCl (182.1 mg, 0.95 mmol) and NHS (109.3 mg, 0.95 mmol) under N2 for 1 h and subsequently added to a suspension of amine-bearing MWNTs 2 (100 mg; obtained Subheading 3.3.2, step 7) in 3 ml of anhydrous DMF. 2. The resultant mixture is stirred at room temperature for 48 h under N2 and in darkness. 3. After the 48-h incubation period, the black solid product is collected by filtration through a nylon-66 membrane filter and washed thoroughly with DMF to remove the excess d-HCPT, EDC·HCl, and NHS. 4. The removal of impurities is monitored by TLC using CHCl3/ CH3OH (90/10, v/v) as the eluent. The TLC plate is visualized with a UV lamp and iodine vapor. 5. After drying under vacuum, the HCPT–MWNT conjugates 3 are obtained as a black solid. Figure  8 shows a representative SEM image of the HCPTMWNTs 3, which are quite soluble in aqueous medium (as indicated by the inset to Fig. 8). We have observed that the amount of remnant amino groups after the conjugation reaction (as determined by the quantitative Kaiser test) is ~0.13 mmol/g of sample, and thus the reaction yield of d-HCPT with MWNT-bound amino groups can be calculated to be ~75%. +

O

NH3

O

O O

O

NH

d-HCPT

O

NH

O

NH

O

HN

O +

H3N

O

MWNTs 2

OHO

N

O

N O

EDC×HCl, NHS, Et3N, dry DMF

HN

O

O

O

O H2N

O

MWNTs 3

Fig. 7. Conjugation of MWNTs to HCPT via amidation reaction between activated d-HCPT and nanotube-bound amino groups.

576

Wu and Jiang

Fig. 8. SEM image of MWNTs conjugated to HCPT (MWNTs 3). The inset shows a photograph of a solution of MWNTs 3 in saline. Reproduced with permission from (20) © 2009 American Chemical Society.

O O O

NH

O

O

O

OH O N

O

O

N

O

O

NH

O

NH

O

O

O

OHO N

O

N O

NH

FITC, dry DMF

HN

O

HN

O H 2N

S

O

MWNTs 3

HN N H

O

O O

HOOC O

O

OH

MWNTs 4

Fig. 9. Labeling of MWNT–HCPT conjugates 3 with fluorescein isothiocyanate (FITC).

3.5. Labeling MWNT–HCPT Conjugates with Fluorescein Isothiocyanate

The remnant amino groups on the MWNTs after conjugation to HCPT (Subheading 3.4) can be used to label the MWNT–HCPT conjugates 3 (Fig. 7) with either fluorescent or radioactive molecules (see Subheading 3.6) that are very useful for tracking the distribution of the nanodrug within cells or inside whole living bodies, respectively (Fig. 9). 1. MWNTs 3 (10 mg) and FITC (2.7 mg) are dispersed in 2 ml of anhydrous DMF. 2. The resulting mixture is stirred at room temperature overnight in darkness. 3. The reaction products are thoroughly rinsed with methanol and diethyl ether via centrifugation. The removal of excess

577

A Practical Strategy for Constructing Nanodrugs

FITC is monitored by TLC using DCM/CH3OH (80/20, v/v) as the eluent. The TLC plate is visualized with a UV lamp. 4. The rinsed FITC-labeled MWNT–HCPT conjugate 4 (Fig. 9) is then dried at room temperature under vacuum to obtain a black solid. 3.6. Labeling MWNT– HCPT Conjugates with the Radioactive Nuclide Technetium99m ( 99mTc) 3.6.1. Synthesis of Diethylenetriamine­ pentaacetic Acid Dianhydride (Fig. 10)

To covalently attach diethylenetriaminepentaacetic acid (DTPA, a general chelator) to MWNTs at a high yield, DTPA dianhydride is initially synthesized by the following procedure: 1. To 10 g (25.4 mmol) of DTPA are added 11 ml (113.8 mmol) of acetic anhydride and 13 ml of dry pyridine. The resulting suspension is stirred at 65°C for 24 h (see Note 7). 2. After cooling down to room temperature, the precipitate yielded is collected by filtration and successively washed with acetic anhydride (3 × 5 ml), dry DMF (3 × 5 ml), and diethyl ether (3 × 5 ml). 3. After drying under vacuum, pure DTPA dianhydride is obtained as a white powder (8.6 g, 24.13 mmol, 95%). Characterization of DTPA by 1H NMR (500 MHz, DMSO-d6) d: 2.60 (t, 4H), 2.76 (t, 4H), 3.32 (s, 2H), 3.72 (s, 8H), 11.76 (s, 1H). 13C NMR (125 MHz, DMSO-d6) d: 50.70, 51.69, 52.63, 54.60, 165.85, 171.88.

3.6.2. Covalent Modification of MWNTs with DTPA (Fig. 11)

DTPA is linked to MWNTs 3 (Fig. 7) via the reaction of remnant nanotube-bound amino groups with DTPA dianhydride. 1. DTPA dianhydride (23.2  mg) is added to a solution of MWNTs 3 (50 mg) in 2 ml of anhydrous DMSO containing 20 ml of Et3N. 2. After sonication for a few minutes, the resulting mixture is stirred at room temperature under darkness overnight. 3. The crude product is collected by filtration through a nylon66 membrane filter (200-nm pore size) and washed thoroughly with DMSO to remove the excess DTPA dianhydride. 4. The removal of excess regents is monitored by TLC using isopropanol/ethyl acetate/water (7/7/5, v/v/v) as the ­eluent. The TLC plate is visualized with iodine vapor.

HO O

OH

O N OH

OH

O

N O

N OH

OH O

acetic anhydride dry pyridine

O

N

O O

Fig. 10. Synthesis of diethylenetriaminepentaacetic acid (DTPA).

N

O

O

N

O O

578

Wu and Jiang

O O O

NH

O

O

OH O

N

O

O

O

N

O

O

NH

O

DTPA dianhydride

NH O

NH

O

HN

O H 2N

O

N OH

MWNTs 3

O

N O

O

N

O

O

OH

HO O

O

O

OH O N O

Et3N, dry DMSO HN

O

N OH

H N

O

O

MWNTs 8

Fig. 11. Conjugation of MWNTs 3 to DTPA.

5. After drying under vacuum, the DTPA-modified MWNTs 8 (Fig. 11) are obtained as a black solid. 3.6.3. Radiolabeling of DTPA-Modified MWNTs 8 with 99mTc (Fig. 12)

Tc (T1/2 = 6.02 h, Eg = 141 keV) is selected to label the MWNT– HCPT conjugates due to its stable nature and appropriate radiant energy range.

99m

1. To a solution of DTPA-modified MWNTs 8 (Fig.  11) in phosphate-buffered saline (pH 7.4, 0.1  M) are added stannous chloride and 99mTcO4−. The resulting mixture is stirred for 20 min at room temperature. 2. Afterward, the 99mTc-labeled MWNT-HCPT product (MWNTs 5) is obtained by filtration through a nylon-66 membrane filter (pore size 200 nm). 3. The radiochemical yield is determined by paper chromatography using Whatman No. 1 chromatography paper strips (1.5 × 15 cm): In detail, the test solution is applied at a position located 1.5 cm from the lower end of the strip. The sample is then chromatographed in 0.9% saline until the solvent reaches the top of the strip. After drying and subsequently cutting the strip into equal segments (1  cm), the level of radioactivity is measured using a NaI(Tl) scintillator. TcO4− and 99mTc-labeled MWNTs show Rf values of 0.9 and 0 in saline, respectively. Our results suggest that the radiochemical purity of the synthesized 99mTc-labeled MWNTs 5 (Fig.  12) is higher than 99%. Moreover, no change in purity is detected after keeping the samples in a saline solution for 72 h, indicating their high stability. After the MWNT–HCPT conjugates have been successfully labeled with radioactive nuclide 99mTc, the in vivo behavior of the radiolabeled conjugates can be traced by using single-photon 99m

A Practical Strategy for Constructing Nanodrugs

O O O

NH

O

N OH

O

O

O

N

O O

stannous chloride,99mTcO4−

N OH

H N

O

O

O

O

O

OHO

N

O

N O

O

HN

O

O O O

N O

MWNTs 8

NH

NH

O

O

N

OHO N O

OH

O

O

NH

HN HO

O

O

579

O

ON 99mTc

O

O

H N

N

O

O O

MWNTs 5

Fig. 12. Labeling of DTPA-modified MWNT–HCPT conjugates 8 with radioactive nuclide technetium-99m (99mTc).

Fig. 13. (a) SPECT images of a hepatic H22 tumor-bearing mouse at various time points after injection of radiolabeled MWNTs 5 via the tail vein under anesthetized conditions. The arrows denote the regions displaying the tumor. (b) SPECT images of a tumor-free rat at various time points after injection of MWNTs 5 via the tail vein under anesthetized conditions.

emission computed tomography (SPECT) imaging. For example, the biodistribution of MWNTs 5 in both subcutaneous hepatic H22 tumor-bearing mice and tumor-free rats at various time points post-injection of MWNTs 5 via the tail vein can be observed in the SPECT images shown in Figs. 13a, b, respectively. In particular, a significant accumulation of MWNTs 5 in tumor tissue is seen in the mouse (Fig. 13a). In other tissues, a similar distribution of MWNTs 5 in the mouse and rat is also revealed by the SPECT images.

580

Wu and Jiang

4. Notes 1. Since there are two hydroxyl groups (at C-10 and C-20) in each HCPT molecule, two corresponding kinds of esters should be yielded by the reaction between succinic anhydride and HCPT. It has been well demonstrated that the 10-hydroxy group in HCPT can be esterified with yields of 40–60% in an aprotic polar solvent, such as DMF or DMSO, whereas the 20-hydroxy group is not as easily esterified under such condition even if 3.0  eq of acylating agents are used. The lower reactivity at the C-20 position arises because a strong hydrogen bond is readily formed between the 20-hydroxy and 21-carbonyl groups in polar solvents such as DMF, which greatly hinders the combination between the 20-hydroxy group and acylating agents (19). 2. To suppress the yield of the bis-protected diaminotriethylene glycol compound, an excess amount of diaminotriethylene glycol and dropwise addition of di-tert-butyl dicarbonate are required. 3. Both the power of the ultrasonic generator and the relative position of the flask inside the water bath can greatly influence the effectiveness of the ultrasonic treatment, and consequently the length of the obtained oxidized MWNTs and the amount of carboxylic acid groups generated on the oxidized MWNTs. Generally, under the same treatment duration and temperature conditions, the higher the sonicator power output, the shorter the obtained oxidized MWNTs and the greater the number of carboxylic groups generated on the nanotube surface. Additionally, placing the flask at different positions within the water bath sonicator can result in exposure to different intensities of ultrasonic power. 4. The Kaiser test is based on the reaction of free amine functional groups with ninhydrin, leading to the formation of a chromophoric compound whose concentration can be determined using UV–visible spectroscopy. 5. Before use, THF should be dried with sodium under reflux in the presence of benzophenone until a characteristic blue color is evident in the solvent. 6. The structure of the carbon nanotubes (e.g., diameter, and especially the defect content) can greatly influence the amount of carboxylic groups produced by the oxidation treatment. Therefore, different batches of carbon nanotubes may exhibit significantly different loadings of functional groups even if they have been exposed to exactly the same oxidation reaction conditions.

A Practical Strategy for Constructing Nanodrugs

581

7. The DTPA dianhydride synthesis reaction should be performed in a round-bottom flask equipped with a reflux condenser bearing a CaCl2 drying tube or an oil bubbler to eliminate atmospheric moisture.

Acknowledgments This research project was supported by the Natural Science Foundation of China (Nos. 50802040, 50625311, and 20874042) and the Ph.D. Programs Foundation of the Ministry of Education of China (No. 200802841037). References 1. Prokop, A., And Davidson, J. (2008) Nanovehicular intracellular delivery systems. J. Pharm Sci. 97, 3518–3590. 2. Lu, F., Gu, L., Meziani, M., Wang, X., Luo, P., Veca, L., Cao, L., And Sun, Y. (2009) Advances in bioapplications of carbon nanotubes. Adv. Mater. 21, 139–152. 3. Liu, Z., Chen, K., Davis, C., Sherlock, S., Cao, Q., Chen, X., And Dai, H. (2008) Drug Delivery with carbon nanotubes for in vivo cancer treatment. Cancer Res. 68, 6652–6660. 4. Ali-Boucetta, H., Al-Jamal, K., Mccarthy, D., Prato, M., Bianco, A., And Kostarelos, K. (2008) Multiwalled carbon nanotube-doxorubicin supramolecular complexes for cancer therapeutics. Chem. Commun. 459–461. 5. Bhirde, A., Patel, V., Gavard, J., Zhang, G., Sousa, A., Masedunskas, A., Leapman, R., Weigert, R., Gutkind, J., And Rusling, J. (2009) Targeted Killing of cancer cells in vivo and in vitro with EGF-directed carbon nanotube-based drug delivery, ACS Nano 3, 307–316. 6. Chen, J., Chen, S., Zhao, X., Kuznetsova, L., Wong, S., And Ojima, I. (2008) Functionalized single-walled carbon nanotubes as rationally designed vehicles for tumor-targeted drug delivery. J. Am. Chem. Soc. 130, 16778–16785. 7. Welsher, K., Liu, Z., Daranciang, D., And Dai, H. (2008) Selective Probing and imaging of cells with single walled carbon nanotubes as near-infrared fluorescent molecules, Nano Lett. 8, 586–590. 8. Liu, Z., Li, X., Tabakman, S., Jiang, K., Fan, S., And Dai, H. (2008) Multiplexed multicolor Raman imaging of live cells with isotopically

modified single walled carbon nanotubes. J. Am. Chem. Soc. 130, 13540–13541. 9. Cherukuri, P., Bachilo, S., Litovsky, S., And Weisman, R. (2004) Near-Infrared fluorescence microscopy of single-walled carbon nanotubes in phagocytic cells. J. Am. Chem. Soc. 126, 15638–15639. 10. Cherukuri, P., Gannon, C., Leeuw, T., Schmidt, H., Smalley, R., Curley, S., And Weisman, R. (2006) Mammalian pharmacokinetics of carbon nanotubes using intrinsic near-infrared fluorescence. Proc. Natl Acad. Sci. USA 103, 18882–18886. 11. Gannon, C., Cherukuri, P., Yakobson, B., Cognet, L., Kanzius, J., Kittrell, C., Weisman, R., Pasquali, M., Schmidt, H., Smalley, R., And Curley, S. (2007) Carbon Nanotube-enhanced thermal destruction of cancer cells in a noninvasive radiofrequency field, Cancer 110, 2654–2665. 12. Pastorin, G., Wu, W., Wieckowski, S., Briand, J., Kostarelos, K., Prato, M., And Bianco, A. (2006) Double Functionalisation of carbon nanotubes for multimodal drug delivery. Chem. Commun. 1182–1184. 13. Wu, W., Wieckowski, S., Pastorin, G., Benincasa, M., Klumpp, C., Briand, J., Gennaro, R., Prato, M., And Bianco, A. (2005) Targeted delivery of amphotericin b to cells by using functionalized carbon nanotubes. Angew. Chem. Int. Ed. Engl. 44, 6358–6362. 14. Schipper, M., Nakayama-Ratchford, N., Davis, C., Kam, N., Chu, P., Liu, Z., Sun, X., Dai, H., And Gambhir, S. (2008) A Pilot toxicology study of single-walled carbon nanotubes in a small sample of mice. Nat. Nanotechnol. 3, 216–221.

582

Wu and Jiang

15. Yang, S., Wang, X., Jia, G., Gu, Y., Wang, T., Nie, H., Ge, C., Wang, H., And Liu, Y. (2008) Long-term accumulation and low toxicity of single-walled carbon nanotubes in intravenously exposed mice, Toxicol. Lett. 181, 182–189. 16. Liu, Z., Cai, W., He, L., Nakayama, N., Chen,  K., Sun, X., Chen, X., And Dai, H. (2007) In Vivo biodistribution and highly efficient tumour targeting of carbon nanotubes in mice. Nat. Nanotechnol. 2, 47–52. 17. Banerjee, S., Hemraj-Benny, T., And Wong, S. (2005) Covalent surface chemistry of singlewalled carbon nanotubes Adv. Mater. 17, 17–29.

18. Sarin, V. K., Kent, S. B. H. Tam, J. P. Merrifield, R. B. (1981) Quantitative monitoring of solidphase peptide synthesis by the ninhydrin reaction. Anal. Biochem. 117, 147–157. 19. Pan, X., Han, R., And Sun, P. (2003) Regioselective synthesis and cytotoxicities of camptothecin derivatives modified at the 7-, 10- and 20-positions. Bioorg. Med. Chem. Lett. 13, 3739–3741. 20. Wu, W., Li, R., Bian, X., Zhu, Z., Ding, D., Li, X., Jia, Z., Jiang, X., and Hu, Y. (2009) Covalently combining carbon nanotubes with anticancer agent: preparation and antitumor activity. ACS Nano 3, 2740– 2750.

Chapter 36 Design and Synthesis of Biofunctionalized Metallic/Magnetic Nanomaterials Eun-Kyung Lim, Seungjoo Haam, Kwangyeol Lee, and Yong-Min Huh Abstract Organic solvent-soluble nanocrystals suitable for magnetic resonance imaging are prepared by two routes, namely, a coprecipitation method and a multiple-step thermal decomposition method (seed-mediated growth). The size, shape, crystallinity, phase, and composition of the prepared nanocrystals are determined by various characterization techniques, including transmission electron microscopy, vibratingsample magnetometer, X-ray diffraction, and inductively coupled plasma mass spectroscopy. Subsequently, the organic-soluble nanocrystals are rendered water-soluble by two methods, the microemulsion method and the ligand exchange method, for biomedical applications. Detailed protocols for the preparation of water-soluble nanocrystals, as well as procedures for drug-loading and antibody conjugation to the watersoluble nanocrystals are provided. Key words: Magnetic nanocrystals, Magnetic nanoparticles, Microemulsion, Ligand exchange, Antibody, Drug, Coprecipitation, Seed-mediated growth

1. Introduction Colloidal nanocrystals can be utilized in various biomedical applications, including separation, diagnosis, and therapy. In particular, magnetic nanoparticles show great potential for applications in magnetic resonance imaging (MRI), which is a powerful tool in medicine due to its noninvasiveness, excellent spatial resolution, and tomographic imaging capabilities with a deep tissue penetration depth (1–3). For application as MRI contrast agents, a high magnetic sensitivity of the magnetic nanocrystals has to be guaranteed, which requires homogeneity of nanoparticle size and composition. In general, two methods are available for the synthesis of magnetic nanocrystals (1): coprecipitation and thermal decomposition. The coprecipitation of suitable cations and anions under Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2_36, © Springer Science+Business Media, LLC 2011

583

584

Lim et al.

controlled pH can produce a large amount of magnetic nanocrystals that are readily soluble in aqueous solutions. However, the control of particle size is notoriously difficult with this method (4, 5). An alternative emerging method is to decompose a suitable metal precursor in organic solvents in the presence of surfactants. This method provides nanoparticles with high magnetization (emu/g) and affords excellent control over particle size and composition; moreover, the process can be scaled-up for mass production (1, 3, 6). However, for practical applications in the biomedical fields, magnetic nanocrystals should attain good solubility in aqueous phase. To increase the solubility of organicsoluble nanoparticles in aqueous solutions, the surface of the nanoparticles can be modified by water-soluble moieties. Two major strategies commonly used to achieve this are the following: (1) exchanging hydrophobic surfactants on the magnetic nanocrystals with hydrophilic agents (ligand exchange method) and (2) wrapping magnetic nanocrystals using amphiphilic agents (microemulsion/addition method). The exchange method is based on replacing hydrophobic surfactants on the magnetic nanocrystals with hydrophilic agents containing polar groups, such as poly(acrylic acid) (PAA), poly(allylamine), and poly(sodium styrene sulfonate) (7). The microemulsion/addition method requires the dispersion of organic-soluble nanoparticles in an organic solvent, whose microdroplets are then dispersed in an aqueous phase containing amphiphilic polymers, followed by removal of the organic solvent by evaporation (8–14). This method is particularly advantageous because various other moieties such as drug molecules or imaging agents can be loaded within the polymer shell. The polymers usually employed in this approach include poly(ethylene glycol)-based amphiphilic copolymers, poly(ethylene glycol)–block poly(propylene glycol)– block poly(ethylene glycol) (PEG–PPG–PEG, commercial name Pluronic® F127), and poly(ethylene glycol)–poly(lactic acid) (PEG-PLA). These polymers provide for both sufficient colloidal stability under physiological conditions for long circulation times, and a high level of biocompatibility. Finally, the surface of the polymer shell can also be decorated by targeting moieties, namely antibodies (e.g., anti-epidermal growth factor, anti-vascular endothelial growth factor, Herceptin®, Erbitux®) or aptamers, for specific targeting of various lesions (3, 15–17).

2. Materials 2.1. Equipment

1. Ultrasonicator. 2. Agitator equipped with a Teflon impeller. 3. Laser light scattering apparatus (ELS-Z, Otsuka Electronics).

Design and Synthesis of Biofunctionalized Metallic/Magnetic Nanomaterials

585

4. Thermogravimetric analyzer (TA Instruments, New Castle, DE). 5. High-resolution transmission electron microscope. 6. Scanning electron microscope. 7. Centrifuge. 8. UV–visible spectrophotometer. 9. Microplate absorbance spectrophotometer. 10. X-ray diffractometer (Rigaku). 11. Vibrating sample magnetometer (Lakeshore, Westerville, OH). 12. Heating mantle and thermocouple. 13. Gel filtration equipment. 14. Lyophilizer. 15. Fume hood. 16. Permanent magnet (0.5 T). 2.2. Reagents

1. Ferric chloride tetrahydrate. 2. Ferrous chloride, anhydrous. 3. Ammonium hydroxide solution. 4. Hydrogen peroxide. 5. Hydrochloric acid. 6. Ammonium thiocyanate. 7. Iron (III) acetylacetonate (Fe(acac)3). 8. Manganese (II) acetylacetonate (Mn(acac)2), anhydrous. 9. 1,2-Hexadecanediol. 10. Oleic acid. 11. Oleylamine. 12. Dodecanoic acid. 13. Dodecylamine benzyl ether. 14. Hexane. 15. Toluene. 16. Ethanol. 17. Poly(acrylic acid). 18. Diethylene glycol (DEG). 19. 1-Ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDC). 20. Sulfo-N-hydroxysulfosuccinimide (sulfo-NHS) (Pierce). 21. Phosphate-buffered saline (PBS), pH 7.4 (Gibco®). 22. Herceptin® (Roche). 23. Erbitux® (Merck). 24. BCA protein assay kit (Pierce).

586

Lim et al.

25. Sephacryl® S-300 HR column. 26. Centricon®, Centriprep®, and Amicon Ultra® centrifugal filter units (Millipore). 27. Dimethyl sulfoxide (DMSO).

3. Methods 3.1. Synthesis of Magnetic Nanocrystals by the Coprecipitation Method

Magnetic g-Fe2O3 nanocrystals (MNCs) with average diameters between 6 and 13  nm, formed by the coprecipitation of metal ions by a base (NaOH or NH3 · H2O) in aqueous solution, can be stably dispersed in organic phase. The synthesis procedure involves the following steps: 1. All glassware and stirring devices used in this preparation were cleaned thoroughly to avoid any contamination during the experiment (see Note 1). 2. 6.0 g of ferric chloride tetrahydrate (FeCl2·4H2O) and 7.3 g of anhydrous ferrous chloride (FeCl2) were mixed in 20 mL of ultra pure water in a 500-mL glass beaker (see Note 3). 3. At room temperature, the mixed solution was mechanically stirred for 30 min at 800 rpm using an agitator equipped with a Teflon impeller (see Note 4). 4. Subsequently, 20  mL of ammonia hydroxide solution was rapidly injected into the reaction mixture, and further stirred for 30 min at 800 rpm. Following this, the reaction mixture was heated to 70°C for 30 min (see Note 6). 5. 2.0 g of dodecanoic acid and 3.2 mL of oleic acid were added into the reaction mixture, and the resulting solution was stirred for 30 min at 1,000 rpm. 6. The gelatinous precipitates were collected by a 0.5-T permanent magnet and washed with ultra pure water. 7. The shape and size of the synthesized hydrophobic magnetic nanoparticles were characterized by transmission electron microscopy (TEM) (Fig.  1a). Magnetic properties were measured using a vibrating-sample magnetometer (VSM) at 298 K (Fig. 1b); in addition, the phase structure of the magnetic nanoparticles was determined by analyzing their X-ray diffraction patterns (2q: 30.1°, 35.4°, 43.1°, 56.9°, 57.5°, and 62.5° at 298 K). 8. The total iron content within the magnetic nanoparticles was evaluated by using a spectrophotometric method. 200 mL of magnetic nanoparticles were mixed with 0.5  mL of HCl/ H2O2 (2:3, v/v) solution, which induced oxidation of Fe2+ to Fe3+. To the solution was added 0.5  mL of 1% ammonium

Design and Synthesis of Biofunctionalized Metallic/Magnetic Nanomaterials

587

Fig. 1.  (a) TEM images of magnetic nanocrystals prepared by the coprecipitation method (Left panel: D0 = 6.5 nm; right panel: D0 = 11 nm), and (b) their magnetization curves (logarithmic scale). Reproduced with permission from (4) © 1998 Materials Research Society.

Fig. 2. Schematic representation of the reaction of Mn(acac)2 and Fe-(acac)3 with surfactants at high temperature to produce monodisperse MnFe2O4 nanocrystals, which can be easily isolated from the reaction by-products and the high-boiling point solvent.

thiocyanate solution, followed by absorption measurement of the thiocyanate complex at l = 480 nm (18). In addition, the concentrations of each ion were measured by inductively coupled plasma atomic emission spectroscopy analysis. 3.2. Synthesis of Magnetic Nanocrystals by the Thermal Decomposition Method 3.2.1. Synthesis of Magnetic MnFe2O4 Nanocrystal Seeds

Magnetic MnFe2O4 nanocrystals with a narrow size distribution and high magnetic sensitivity can be synthesized by the following multiple thermal decomposition method (see Note 5 and Fig. 2):

1. All glasswares and apparatus used in this preparation were cleaned properly to avoid any contamination during the experiment. In particular, to eliminate moisture, a 250-mL threeneck round-bottomed flask and condenser were completely

588

Lim et al.

dried in a vacuum drying oven (100°C) under reduced pressure (0.1 mmHg) (see Note 1). 2. For the synthesis of highly monodisperse magnetic nanocrystals, water should be excluded from the system (see Note 2). Thus, the three-neck flask was purged with high-purity nitrogen (N2) gas to remove moisture. 3. 706.3  mg of iron (III) acetylacetonate, 253.2  mg of anhydrous manganese (II) acetylacetonate, and 2.6  g of 1,2-­hexadecanediol were added into the three-neck roundbottomed flask. Ligand pairs of either 1.2  g of dodecanoic acid/1.1 g of dodecylamine or 1.7 g of oleic acid/1.6 g of oleylamine were introduced into the flask, depending on the desired types of MNCs to be prepared. 4. 20 mL of benzyl ether was added into the flask, and the reaction mixture was placed under vacuum, followed by N2 flow. All work should be performed in a fume hood (see Note 4). 5. The flask was heated to 100°C for 1 h and then, under a blanket of N2, further heated (300°C) to reflux for 30 min with magnetic stirring (see Note 6). 6. After termination of the reaction, the black-brown mixture was cooled to room temperature by removing the heat source. 40  mL of ethanol was added into the reaction mixture to precipitate the magnetic nanocrystals, which could be separated by centrifugation at 1,000 × g for 10 min. 7. The black precipitate was dissolved in 4 mL of hexane, and 50 mL of oleic acid and 50 mL oleylamine were added into the solution. 8. 40 mL of ethanol was added again to reprecipitate the magnetic nanocrystals, and the precipitates were collected by centrifugation at 1,000 × g for 10 min. 9. The precipitated magnetic nanocrystals were redispersed into hexane.

Fig. 3. TEM images of (a) 6 nm, (b) 10 nm, (c) 12 nm Fe3O4 nanocrystals, and (d) 12 nm MnFe2O4 nanocrystals deposited from their hexane dispersions on an amorphous carbon-coated copper grid and dried at room temperature. Reproduced with permission from (1) © 2003 American Chemical Society and from (3) © 2007 Nature Publishing Group.

Design and Synthesis of Biofunctionalized Metallic/Magnetic Nanomaterials

589

10. The shape and size of the magnetic nanocrystal seeds were characterized by TEM (Fig. 3a). 3.2.2. Synthesis of 10 nm Magnetic Nanocrystals via Seed-Mediated Growth

1. 706.3  mg of iron (III) acetylacetonate, 253.2  mg of anhydrous manganese (II) acetylacetonate and 2.6 g of 1,2-­ hexadecanediol were added into a three-neck round-­bottomed flask. Ligand pairs of either 0.4 g of dodecanoic acid/0.4 g of dodecylamine or 0.6 g of oleic acid/0.5 g of oleylamine were introduced into the flask, depending on the desired types of MNCs to be prepared. 2. 100 mg of 4 nm magnetic nanocrystal seeds in 4 mL of hexane (from Subheading 3.2.1) were added into the reaction mixture, and the reaction mixture was magnetically stirred under a flow of N2. 3. To remove the hexane, the reactants were heated to 100°C for 1  h, and then to 200°C for 2  h under a blanket of N 2, and finally heated to reflux (300°C) for 30  min (see Note 6). 4. Purification of the 10-nm magnetic nanocrystal product was carried out as described in Subheading 3.2.1. 5. The shape and size of the magnetic nanocrystals were characterized by TEM (Fig. 3b).

3.2.3. Synthesis of 12 nm Magnetic Nanocrystals from 10 nm Magnetic Nanocrystal Seeds

1. 706.3  mg of iron (III) acetylacetonate, 253.2  mg of anhydrous manganese (II) acetylacetonate, and 2.6 g of 1,2hexadecanediol were added into a three-neck round-bottomed flask. Ligand pairs of either 0.4 g of dodecanoic acid/0.4 g of dodecylamine or 0.6 g of oleic acid/0.5 g of oleylamine were introduced into the flask, depending on the desired types of MNCs to be prepared. 2. 100  mg of 10  nm magnetic nanocrystal seeds in 4  mL of hexane (from Subheading 3.2.2) were added into the reaction mixture, and the reaction mixture was magnetically stirred under a flow of N2. 3. To remove the hexane, the reactants were heated to 100°C for 1 h, and then to 200°C for 2 h under a blanket of N2, and finally heated to reflux (300°C) for 30 min (see Note 6). 4. Purification of the 12-nm magnetic nanocrystal product was carried out as described in Subheading 3.2.1. 5. The shape and size of the synthesized hydrophobic magnetic nanocrystals were characterized by TEM (Fig. 3c, d). Further detailed characterization of the magnetic nanocrystals produced by the thermal-decomposition method was carried out as outlined above (see Subheading 3.1).

590

Lim et al.

3.3. Preparation of Hydrophilic Magnetic Nanocrystals 3.3.1. Preparation of Hydrophilic Magnetic Nanocrystals by the Nanoemulsion Method (Fig. 4)

1. Hydrophobic magnetic nanocrystals (50 mg) were dissolved in 4 mL of a nonpolar organic solvent such as hexane, chloroform, or toluene. Amphiphilic polymers (100  mg) as stabilizer were dissolved into 20  mL of aqueous (PBS) solution (see Note 1). 2. The magnetic nanocrystal-containing organic solution was introduced into the polymer solution in a100-mL beaker, which spontaneously formed an emulsion. The reaction mixture was then stirred at 1,000 rpm for 10  min at room temperature in an ultrasonicator (water bath type, 190  W) equipped with a Teflon impeller. 3. To evaporate the organic solvent, the mixture was further stirred magnetically for 24  h, and the nanoparticles were purified with two cycles of centrifugation at 2,000 × g for 30  min. The precipitated nanoparticles were redispersed in purified deionized water. 4. The average diameter and zeta-potentials of the prepared hydrophilic magnetic nanocrystals (HMNCs) were measured by laser light scattering (Fig. 5a). The morphology and shape of the nanoparticles were characterized by scanning electron microscopy (SEM) and TEM (Fig. 5b). The magnetic properties were measured using a VSM at 298  K (Fig.  5c). Furthermore, the polymer content and loaded amount of magnetic nanocrystals in the HMNCs were determined by using a thermogravimetric analyzer (TGA) (Fig. 5d).

Fig. 4. Schematic illustration of the preparation of magnetic nanoparticles by the nanoemulsion method.

Design and Synthesis of Biofunctionalized Metallic/Magnetic Nanomaterials

591

Fig. 5. (a) Size distributions of magnetic nanocrystals (MNCs, solid line) and hydrophilic magnetic nanocrystals (HMNCs, dashed line) as measured by laser light scattering. (b) TEM image, (c) magnetization curve, and (d) TGA curve acquired for HMNCs. Reproduced with permission from (12) © American Chemical Society, (13) © Elsevier, and (16) © Royal Society of Chemistry.

3.3.2. Preparation of Drug-Loaded Magnetic Nanocrystals by the Nanoemulsion Method

1. Hydrophobic magnetic nanocrystals (10  mg) and drugs (5–10  mg) were dissolved in 4  mL of a nonpolar organic solvent nonmiscible with water. Amphiphilic polymers (100  mg) as stabilizer were dissolved in 20  mL of aqueous solution (see Note 1). 2. The preparation and physical characterization of the drugloaded magnetic nanocrystals (DMNCs) (Fig. 6) were carried out as described in Subheading 3.3.1. 3. For determination of the drug incorporation efficiency, 100 mL of DMNCs was lyophilized. The freeze-dried nanoparticles were dissolved in 4 mL of DMSO, and the solution was measured with a UV–visible spectrophotometer. The drug content (% w/w) and drug entrapment efficiency (%) are represented by Eqs. 1 and 2, respectively: Amount of drug in nanoparticles



Drug content (% w/w) =



Drug entrapment efficiency (%) =

Amount of nanoparticles

´ 100 (1)

Amount of drug in nanoparticles ´ 100 (2) Initial amount of drug used

592

Lim et al.

Fig. 6. TEM images of drug-loaded hydrophilic magnetic nanocrystals (DMNCs) containing (a) MnFe2O4 and (b) Fe3O4. (c) Size distributions of DMNCs containing MnFe2O4 (solid line) and Fe3O4 (dashed line). (d) Magnetic hysteresis loops of magnetic nanocrystals (solid line) and DMNCs (dashed line); MnFe2O4 (black lines) and Fe3O4 (gray lines). Reproduced with permission from (15) © Wiley-VCH Verlag GmbH & Co. KGaA.

3.3.3. Preparation of Hydrophilic Magnetic Nanocrystals by the Ligand Exchange Method (Fig. 7a)

1. A solution containing 0.5 g of PAA in 8.0 mL of DEG was prepared in a three-neck round-bottomed flask, and then heated to 110°C with vigorous magnetic stirring under N2 gas flow (see Notes 1 and 4). 2. 1.0  mL of hydrophobic magnetic nanocrystals dispersed in an organic solvent (100  mg/mL) was injected into the hot solution prepared in step 1, which became turbid immediately. 3. Reactants were heated to 240°C and kept for 1 h until the solution became a transparent brownish color (see Note 6). 4. After termination of the reaction, the brownish transparent mixture was cooled to room temperature by removing the heat source. 5. To purify the HMNCs, 20 mL of ethanol or purified water was added to the reaction mixture. The precipitated nanoparticles were separated by centrifugation at 1,000 × g for 15 min. The black precipitates were washed three times with ethanol or purified water and redispersed into purified deionized water. The resulting HMNCs can be well dispersed in aqueous solution. 6. HMNCs produced by the ligand exchange method were characterized in a similar analysis as previously described (see Subheading  3.3.1). The size and shape of the synthesized HMNCs were analyzed by TEM, which indicated a monodisperse size distribution without aggregation (Fig. 7b). Changes in the ratio of the ligand content and the amount of magnetic nanocrystals in the HMNCs were determined by using a TGA (Fig. 7c).

Design and Synthesis of Biofunctionalized Metallic/Magnetic Nanomaterials

593

Fig. 7. (a) Schematic illustration of the general principle of the ligand exchange approach for the preparation of hydrophilic magnetic nanoparticles by using oleic acid-coated iron oxide nanocrystals as a model system. Ligand exchange with PAA in a polyol solvent such as DEG at high temperature renders these nanocrystals water-soluble. (b) TEM images of g-Fe2O3 nanocrystals after ligand exchange. Scale bar = 20 nm. (c) TGA curves of g-Fe2O3 nanocrystals before and after ligand exchange. The measurements were performed under a N2 atmosphere. Reproduced with permission from (7) © American Chemical Society.

3.4. Preparation of Smart AntibodyConjugated Magnetic Nanoparticles

1. Redisperse the HMNCs containing carboxyl groups (prepared in Subheading  3.3 above) to achieve a total polymer concentration of ~1–10 mg/mL in PBS (pH 7.4). 2. The antibody (e.g., Herceptin®, Erbitux®, etc.) to be coupled was dissolved in PBS (see Note 7). 3. The solution of HMNCs prepared in step 1 was added to the antibody solution to give at least a 100-fold molar excess of the amount of antibody. 4. EDC and sulfo-NHS were added to the above solution in equimolar amounts. The absolute amounts of these reagents were at least a tenfold molar excess over the amount of added antibody. After mixing, the solution was further incubated for 6 h at 4°C (see Note 7). 5. The antibody-conjugated magnetic nanoparticles were purified in several ways. Free, unconjugated antibody and antibody-conjugated magnetic nanoparticles were separated by column chromatography using a high-resolution gel filtration column packed with Sephacryl® S-300 HR (see Note 8). Alternatively, centrifugation or centrifugal filter devices

594

Lim et al.

(e.g., Centricon®, Centriprep® or Amicon Ultra®) with ­membranes of the appropriate molecular weight cutoff can also be used. 6. The amount of antibody conjugated to the magnetic nanoparticles was evaluated by using a BCA protein assay kit (see Note 9).

4. Notes 1. All glassware should be cleaned and stored in an oven at 125°C for at 12 h before use to avoid contamination during the experiment. 2. This step is necessary to remove moisture from the experimental apparatus. 3. All reagents and other chemicals are of analytical grade and should be kept in a container tightly closed in a dry and wellventilated place. 4. Caution: Most of the chemicals and reagents are hazardous and should be handled with care in a fume hood. 5. Synthesis of magnetic nanocrystals by the seed-mediated growth method should be conducted under a N2 atmosphere. 6. Use gloves at all times when handling hot glassware or hazardous chemicals. 7. EDC should be stored in a cool place (recommended storage temperature: −20°C) and protected from moisture. In particular, when the reagents are removed from −20°C freezer and placed at room temperature, collected moisture around the reagent bottles should be wiped before opening the bottle. Sulfo-NHS and lyophilized antibody proteins should be stored at 4°C and protected from moisture. 8. To remove undesirable contaminants and to activate dried beads in the column, the Sephacryl® column can be sanitized by running at least three column volumes of 0.1N NaOH solution before loading the nanoparticles. 9. To measure protein concentration using the BCA protein assay, insert 25 ml of each diluted protein standard solution (e.g., bovine serum albumin solution) or experimental sample solution replicate into a 96-well microplate, respectively. Add 200  ml of the BCA working reagent (Reagent A:B = 50:1, v/v) to each well, and mix thoroughly on a plate shaker for 30  s. Cover the microplate and incubate at 37°C. After 30  min, cool the microplate to room temperature. Finally, measure the absorbance at l = 562  nm using a microplate spectrophotometer.

Design and Synthesis of Biofunctionalized Metallic/Magnetic Nanomaterials

595

References 1. Sun, S., Zeng, H., Robinson, D. B., Raoux, S., Rice, P. M., Wang, S. X., and Li, G. (2003) Monodisperse MFe2O4 (M = Fe, Co, Mn) nanoparticles. J. Amer. Chem. Soc. 126, 273–279. 2. Huh, Y. M., Jun, Y. w., Song, H. T., Kim, S., Choi, J. s., Lee, J. H., Yoon, S., Kim, K. S., Shin, J. S., Suh, J. S., and Cheon, J. (2005) In vivo magnetic resonance detection of cancer by using multifunctional magnetic nanocrystals. J. Am. Chem. Soc. 127, 12387–12391. 3. Lee, J.-H., Huh, Y.-M., Jun, Y.-w., Seo, J.-w., Jang, J.-t., Song, H.-T., Kim, S., Cho, E.-J., Yoon, H.-G., Suh, J.-S., and Cheon, J. (2007) Artificially engineered magnetic nanoparticles for ultra-sensitive molecular imaging. Nat. Med. 13, 95–99. 4. S. Lefebure, E. D., Cabuil, V., Neveu, S. and Massart, R. (1998) Monodisperse magnetic nanoparticles: Preparation and dispersion in water and oils. J. Mater. Res. 13, 2975–2981. 5. Massart, R. (1981) Preparation of aqueous magnetic liquids in alkaline and acidic media. IEEE Trans. Magn. 17, 1247–1248. 6. Laurent, S., Forge, D., Port, M., Roch, A., Robic, C., Vander Elst, L., and Muller, R. N. (2008) Magnetic iron oxide nanoparticles: Synthesis, stabilization, vectorization, physicochemical characterizations, and biological applications. Chem. Rev. 108, 2064–2110. 7. Zhang, T., Ge, J., Hu, Y., and Yin, Y. (2007) A general approach for transferring hydrophobic nanocrystals into water. Nano Lett. 7, 3203–3207. 8. Solans, C., Izquierdo, P., Nolla, J., Azemar, N., and Garcia-Celma, M. J. (2005) Nanoemulsions. Curr. Opin. in Colloid Interface Sci. 10, 102–110. 9. Lee, S.-J., Jeong, J.-R., Shin, S.-C., Kim, J.-C., Chang, Y.-H., Chang, Y.-M., and Kim, J. D. J.-D. (2004) Nanoparticles of magnetic ferric oxides encapsulated with poly(D,L latide-coglycolide) and their applications to magnetic resonance imaging contrast agent. J. Magn. Magn. Mater. 272–276, 2432–2433. 10. Lee, J., Yang, J., Seo, S.-B., Ko, H.-J., Suh, J.-S., Huh, Y.-M., Haam, S., (2008) Smart nanoprobes for ultrasensitive detection of breast cancer via magnetic resonance ­imaging. Nanotechnology 19, 485101.

11. Lim, E.-K., Yang, J., Park, M.-y., Park, J., Suh, J.-S., Yoon, H.-G., Huh, Y.-M., and Haam, S. (2008) Synthesis of water soluble PEGylated magnetic complexes using mPEG-fatty acid for biomedical applications. Colloids Surf. B: Biointerfaces 64, 111–117. 12. Yang, J., Lee, T.-I., Lee, J., Lim, E.-K., Hyung, W., Lee, C.-H., Song, Y. J., Suh, J.-S., Yoon, H.-G., Huh, Y.-M., and Haam, S. (2007) Synthesis of ultrasensitive magnetic resonance contrast agents for cancer imaging using PEG-fatty acid. Chem. Mater. 19, 3870–3876. 13. Eun-Kyung Lim, J. Y., Jin-Suck Suh, YongMin Huh, Seungjoo Haam. (2009) Selflabeled magneto nanoprobes using tri-aminated polysorbate 80 for detection of human mesenchymal stem cells. J. Mater. Chem. 19, 8958–8963. 14. Zaitsev, V. S., Filimonov, D. S., Presnyakov, I. A., Gambino, R. J., and Chu, B. (1999) Physical and chemical properties of magnetite and magnetite-polymer nanoparticles and their colloidal dispersions. J. Colloid Interface Sci. 212, 49–57. 15. Yang, J., Lee, C.-H., Ko, H.-J., Suh, J.-S., Yoon, H.-G., Lee, K., Huh, Y.-M., and Haam, S. (2007) Multifunctional magneto-polymeric nanohybrids for targeted detection and synergistic therapeutic effects on breast cancer. Angew. Chem. Int. Ed. 119, 8992–8995. 16. Yang, J., Lim, E.-K., Lee, H. J., Park, J., Lee, S. C., Lee, K., Yoon, H.-G., Suh, J.-S., Huh, Y.-M., and Haam, S. (2008) Fluorescent magnetic nanohybrids as multimodal imaging agents for human epithelial cancer detection, Biomaterials 29, 2548–2555. 17. Yang, J., Lee, C.-H., Park, J., Seo, S., Lim, E.-K., Song, Y., Suh, J.-S., Yoon, H.-G., Huh, Y.-M., Haam, S., (2007) Antibody conjugated magnetic PLGA nanoparticles for diagnosis and treatment of breast cancer. J. Mater. Chem. 17, 2695–2699. 18. Zaitsev, V. S., Filimonov, D. S., Presnyakov, I. A., Gambino, R. J., and Chu, B. (1999) Physical and chemical properties of magnetite and magnetite-polymer nanoparticles and their colloidal dispersions. J. Colloid Interface Sci. 212, 49–57.

Index A Absorbance spectroscopy................................................ 474 2-Acetonyl–2-deoxy-galactose (C2-keto-Gal)....... 282–284 Adenovirus type 5 (Ad5)...........................57, 59–60, 63, 64 Affinity chromatography protein A.......................................................... 284, 287 protein G.................................................................. 335 AFM. See Atomic force microscopy Alkaline phosphatase (AP)..........................82–93, 124, 525 Alkanethiolates....................................................... 421, 427 Alkyne-azide cycloaddition.......................32, 57–58, 68, 74 Alkyne-TAMRA...................................................57, 62, 63 Alkyne-terminated poly(NIRF)-poly(Glu) polymer..........................................31, 32, 38–41 Amber suppression................................................. 4, 11–12 Aminoacyl-tRNA synthetase (aaRS)...............4–5, 8, 12–13 Aminooxy group synthesis of aminooxy linker..............319, 405, 413–414 Aminosilane............................ 382–383, 385, 386, 388, 389 Anthon assay.................................................................. 324 Antibody conjugation to water-soluble magnetic nanoparticles......................................... 593–594 multiple labeling................................................... 43–53 Anti-cancer drug................................. 30, 67, 242–243, 259 delivery..................................................................... 519 Antisense oligonucleotide-peptide conjugates........ 224, 228 Antisense oligonucleotides..............................209–210, 239 AP. See Alkaline phosphatase Ascorbic acid...................................................360, 368, 369 Asialofetuin.............................................283–285, 288–292 Asn-linked glycan (N-glycan)..........................282, 286–292 Atomic force microscopy (AFM) chemical functionalization, cantilever probe tips...............385, 387, 389, 391, 392, 394, 396, 398 imaging, DNA-block copolymers (DBC) aggregates in air.......................................... 258, 259 aggregates in fluid....................................... 258, 259 Atom transfer radical polymerization (ATRP)................. 41 Avidin������������������������������������������31, 33–37, 45, 47, 53, 478

Avidin-polymer conjugates avidin-poly(NIRF)-poly(Glu) polymer conjugate............................................. 31, 34–37 Azide azide-terminated poly(NIRF)-(Glu) polymer............................................... 31, 37–38 azido-DNA...................................................... 198, 205 2-azidoethyl lactoside................................272, 274–276 azidohomoalanine......................................56, 59–60, 63 azido sugar.................................................................. 63 azido-TTP.........................................196–202, 205, 206 galactose azide.................................................. 168, 191

B Bacillus brevis gramicidin A............................................................. 538 peptide synthetase TycA........................................... 133 Beer–Lambert relation..................................................... 71 Benzyloxycarbonyl-l-glutaminylglycine (Z-QG)....................... 18, 21, 27, 82, 83, 92, 93 Bicinchoninic acid (BCA) assay.................84, 89, 108–109, 113, 121, 213, 219, 312, 315, 319, 324, 585, 594 Biodegradable polymer..................................................... 17 Bionanoelectronics......................................................... 534 Biosensor������������������������ 401–418, 437, 438, 453, 454, 462, 507, 520, 524 Biotin N’-aminooxymethylcarbonylhydrazino-D-biotin (AOB)............................................285, 289, 290 biotin–phosphine.......................................196, 202–207 biotin-terminated DNA............................................. 44 biotin-terminated poly(NIRF)-poly(Glu) polymer..........................................30, 31, 33–36 biotinylated anti–17b-estradiol antibody.................... 45 biotinylated goat anti-mouse antibody (BT-Ab)....................44–45, 47, 49, 50 biotinylated liposome........................................482, 483, 485–487 thioester............................................330, 332–334, 337, 340, 341

Sonny S. Mark (ed.), Bioconjugation Protocols: Strategies and Methods, Methods in Molecular Biology, vol. 751, DOI 10.1007/978-1-61779-151-2, © Springer Science+Business Media, LLC 2011

597

Bioconjugation Protocols 598  Index

  

Bovine serum albumin (BSA) aminooxylated........................................................... 323 azide-modified...........................................32, 33, 39–41 bromoacetyleted........................................................ 322 polymer conjugates......................................... 32, 39–40 BSA. See Bovine serum albumin

C Cancer vaccine carbohydrate-based................................................... 311 Carbohydrate antigen������������������������������������������������������ 310–311, 318 carbohydrate-protein conjugation............................. 311 Carbon nanotube (CNT) covalent attachment, biomolecules to gold nanoparticle-decorated......................... 523–524 covalent coupling, carboxylated CNT with biomolecules......................................... 521–522 preparation metallic nanoparticle/carbon nanotube composites............................................ 517–519 oxidized CNT for covalent functionalization with biomolecules......................................... 515–516 5(6)-Carboxyfluorescein (CBF)............. 272, 274, 277, 479, 480, 483–485 Carrier protein keyhole limpet hemocyanin (KLH).......................... 311 Cationic peptide-lipid (CatLip)............................. 212, 217 CatLip-PNA conjugates................................................ 212 CBF. See 5(6)-Carboxyfluorescein Cell culture adenovirus infection.............................................. 57, 60 cell adhesion..................................................... 422, 435 cell co-culture....................................422, 425, 432–433 live cell imaging........................................................ 561 transfection................................................159, 210, 214 Cell membranes labeling..................................................................... 561 Cell migration.................................................110, 421–436 Cell patterning aldehyde strategy.......................................425, 434–435 oxyamine strategy......................................425, 433–434 Cell-penetrating peptide (CPP)............................. 209–221 Chemical vapor deposition (CVD)................506, 524–525, 535–536, 538, 540–542, 550, 567 Chemiluminescence...................................43–44, 283, 285, 289–292, 478, 486–488 Chemoselective modification chemoselective labeling of DNA.............................. 195 preparation, biomolecular conjugates through chemoselective ligations............................ 67–78 Click chemistry..............................................57–58, 61–62, 270–271 CNT. See Carbon nanotube

Conducting polymer thin films......................438, 440, 446, 447, 449, 450 Copper-catalyzed azide-alkyne cycloaddition (CuAAC)......... 59, 63, 68, 74, 75, 168–169, 171, 183, 190, 191 CPP. See Cell-penetrating peptide CVD. See Chemical vapor deposition Cyclic voltammetry (CV)........................431, 432, 537, 546 Cysteine bioconjugation......................................... 130–141 Cysteine tag (Cys-tag)............................................ 130–141 Cytotoxicity����������������������������159–162, 242–243, 259–260, 438, 507

D DAG. See Diacylglycerol DBC. See DNA block copolymer 3-Deoxy-d-manno-oct–2-ulosonic acid (Kdo)......320, 321, 326 Diacylglycerol (DAG).............................478, 481, 485, 487 Diafiltration.....................................................107–108, 119 Diagnostic imaging.........................................359, 371–373 Diethylenetriaminepentaacetic acid (DTPA).........359, 361, 371–373, 566, 569, 577–579, 581 Dimethylcasein (DMC)..................................19, 22–24, 27 2,4-Dinitrophenylhydrazine (DNPH)........................... 115 1,2-Dipalmitoyl-sn-glycero–3-phosphoethanolamine (DPPE)..........271–274, 276, 359, 360, 454, 537 Dispase�����������������������������������������������������������������������������26 5,5'-Dithiobis(2-nitrobenzoic acid) (DTNB)........... 97–98, 112, 417 DLS. See Dynamic light scattering DMC. See Dimethylcasein DNA delivery using PEI derivatives................................... 162 enzymatic incorporation, azido-TTP into DNA............................................. 197, 205 hybridization....................... 46, 241–242, 247, 259–262 polymerase................................... 84, 195–197, 205, 242 DNA block copolymer (DBC) drug loading, DNA block copolymer nanoparticles..................................247, 259–262 surface functionalization, DBC nanoparticles by hybridization..................................247, 259–262 synthesis, DNA-b-polypropylene oxide (PPO) block copolymer..............................243–244, 247–250 DNA-polymer conjugates...................................... 358–359 DNA-polymer nanoparticles.................................. 240–241 DNPH. See 2,4-Dinitrophenylhydrazine Double immunodiffusion analysis.................................. 324 Doxorubicin (DOX)........................242–243, 247, 259–260 DPPE. See 1,2-Dipalmitoyl-sn-glycero–3phosphoethanolamine DPPE-triphenylphosphine.............................271–274, 276 Dragendorff ’s reagent..................................................... 374

Bioconjugation Protocols 599 Index     

Drug delivery drug-loaded magnetic nanocrystal (DMNC).............................................. 591–592 drug nanocarriers.............................................. 565–566 DTNB. See 5,5’-Dithiobis(2-nitrobenzoic acid) DTPA. See Diethylenetriaminepentaacetic acid Dye/DNA conjugates competitive immunoassay of 17b-estradiol................. 51 Dynamic light scattering (DLS)............. 148, 157, 241, 257, 264, 272, 275–279, 460, 487

E ECM. See Extracellular matrix EDC. See 1-Ethyl–3-(3-dimethylaminopropyl) carbodiimide EGFR. See Epidermal growth factor receptor EIS. See Electrochemical impedance spectra Electrochemical copolymerization.......................... 440, 446 Electrochemical detection...............................447, 449–450 Electrochemical impedance spectra (EIS).............. 447–448 Electrochemical surface activation.......................... 425, 431 Electrode counter (auxiliary).............. 425, 431, 440, 446, 447, 456 reference..... 425, 431, 440, 441, 446, 447, 456, 541, 546 working...............431, 440, 446, 447, 451, 456, 458, 546 Electron beam (e-beam) deposition.......534, 536, 539–540, 543, 544 Electrophoresis agarose gel........................... 92, 137, 155–157, 163, 371 gel retardation assay.................................................. 156 SDS-polyacrylamide gel electrophoresis (SDS-PAGE)................. 5, 8–11, 13, 20, 22–26, 31, 32, 34–37, 39–41, 58, 62–63, 65, 90–91, 109–111, 122, 138, 139, 285, 287, 290–292, 294, 320–323, 325, 335–337, 418 Electroporation....................................................... 335–336 Electrospray deposition (ESD)............................... 493–499 Electrospray ionization mass spectrometry (ESI-MS)....76–77, 99, 111–112, 116, 123, 163, 199, 272, 301, 337–339, 567, 570 ELISA. See Enzyme-linked immunosorbent assay Ellman’s assay...................... 97–98, 108, 114, 115, 117, 124 Enzyme immobilization custom-built chamber............................................... 468 on reactive polymer films.................................. 464–475 Enzyme-linked immunosorbent assay (ELISA)............ 314, 366–367, 473–474, 478, 481, 486, 491, 492, 498–500 Epidermal growth factor receptor (EGFR)............554, 555, 559–562 Escherichia coli....... 4–5, 7–9, 12, 26, 82, 84, 86–89, 92, 133, 134, 137, 140, 282–283, 295, 331, 335, 404, 411, 455, 460–462 ESD. See Electrospray deposition

ESI-MS. See Electrospray ionization mass spectrometry 17b-Estradiol........................................................45, 46, 51 1-Ethyl–3-(3-dimethylaminopropyl) carbodiimide (EDC)............. 117, 148, 152, 153, 403, 406, 408, 410, 414, 440, 446, 451, 509, 521–522, 526, 528, 585, 593, 594 3,4-Ethylenedioxythiophene (EDOT-COOH).....438, 439, 444–446, 449, 450 Extein������������������������������������������������������������� 131–132, 136 Extinction coefficient....................... 14, 116, 138, 177, 191, 205, 263, 574 Extracellular matrix (ECM)................................... 421, 428

F Fast protein liquid chromatography (FPLC)..........7, 10, 31, 34, 35, 39, 40, 106, 107, 245, 246, 252, 256, 555, 561 FCS. See Fluorescence correlation spectroscopy Fibroblast cells 3T3 cells.................... 148, 159–161, 424, 425, 432–434 Fibronectin..................................................................... 428 FITC. See Fluorescein isothiocyanate FITC-b-Ala-QG peptide-tag directed protein labeling, as substrate for microbial transglutaminase............................. 82 FITC-labeled streptavidin.......................................... 45, 47 Flory radius (RF)............................................................. 460 Fluorenylmethyloxycarbonyl (Fmoc)........ 68, 70–73, 76, 77, 83–85, 103, 118, 119, 212, 215, 216, 224, 225, 227, 230, 233, 236, 345, 348, 350, 353, 405, 406, 415, 416, 423, 428 Fluorescein isothiocyanate (FITC).................44, 45, 47, 49, 51–53, 82–86, 90–91, 93, 359, 366, 566, 567, 569, 576–577 Fluorescence...... 3, 13–14, 23, 24, 36, 37, 41, 43–53, 63, 85, 90–92, 104, 108, 121, 122, 148, 149, 151, 159–163, 241, 245, 246, 251, 257–258, 272, 274–275, 285, 291–294, 449, 459, 462–463, 472, 478–480, 482–485, 553, 554, 556, 561 dye leakage assay....................................................... 478 labeling........................................ 44, 285, 286, 292–294 Fluorescence correlation spectroscopy (FCS) autocorrelation function G(tc).................................. 258 Fluorescence recovery after photobleaching (FRAP).... 459, 462–463 Fluorescence resonance energy transfer (FRET).......... 3–15 Fmoc. See Fluorenylmethyloxycarbonyl Folic acid-modified oligonucleotides.............................. 261 FPLC. See Fast protein liquid chromatography FRAP. See Fluorescence recovery after photobleaching Freeze-fracture microscopy..................................... 157–159 FRET. See Fluorescence resonance energy transfer Fusion protein........................ 132, 133, 137–138, 140, 285, 286, 292–295, 334, 407

Bioconjugation Protocols 600  Index

  

G a1,3-Galactosyltransferase (a3Gal-T)................... 283, 290 GAS infection. See Group A streptococcal infection Gene delivery............................56, 145, 160, 269, 270, 369, 370, 534 GFP. See Green fluorescent protein Glucose oxidase-avidin D (GOD-A) enzyme........ 441, 449 Glycoconjugates glycoconjugate vaccine...................................... 309–315 Glycomimetic oligonucleotide conjugates preparation of tetragalactosyl-mannose oligonucleotide conjugates.................... 188–189 Glycopeptides..................................................293, 343–353 Glycoproteins................................. 282, 283, 288, 293, 295, 329–341, 395–397, 399 Glycosylamine.................................344, 346–347, 349–353 Glycosylation anti-HER2 scFv fusion protein.........285, 286, 292–294 Glycosyltransferases b4Gal-T1-Y289L mutant................................. 283, 288 human polypeptide-a-Nacetylgalactosaminyltransferase II (h-ppaGalNAc-T2), AuNP................. 282–283 280 SGG282-a3Gal-T mutant.......................283, 288–290 Glycyl-glycine (Gly-gly) test............................................ 97 GOD-A enzyme. See Glucose oxidase-avidin D enzyme Gold nanoparticle (AuNP)..... 509, 517–519, 522–524, 528, 541, 550 Gold substrates thin film deposition using electron beam evaporation................................................... 423 Green fluorescent protein (GFP)............... 57, 91, 148, 159, 161, 361, 371, 404, 411, 412, 425, 432, 433 GRGDS peptide ketone-terminated............................................ 425, 434 oxyamine-terminated................................................ 425 Group A streptococcal (GAS) infection......................... 298

H Habeeb assay...................................................108, 123–124 HCPT. See 10-Hydroxycamptothecin HeLa pLuc705 cell line.......................................... 210–211 HES. See Hydroxyethyl starch HES 70-amine................................................19, 21–23, 27 HESylation...................................................................... 18 Hexamethylenediamine (HMDA)..................18, 19, 21, 27 HIC. See Hydrophobic interaction chromatography High mannose oligosaccharide............... 344, 345, 347, 349, 351–352 High performance liquid chromatography (HPLC)...................69, 74, 75, 84, 86, 100, 106, 109–111, 113, 116, 118–123, 126, 169, 182, 183, 188, 189, 191, 212, 215, 217, 218, 226, 229, 230, 232, 233, 235, 299, 304, 337, 338,

345, 346, 350, 351, 353, 359, 404, 405, 411, 415, 467, 469, 471–475, 538, 555, 556, 561 HIV gp120, 343–353 HMDA. See Hexamethylenediamine HMNC. See Hydrophilic magnetic nanocrystal HPLC. See High performance liquid chromatography HSA. See Human serum albumin hTSH. See Human thyroid stimulating hormone Human immunoglobulin G subclass 1 fragment crystallizable (IgG1 Fc).........330–332, 334–341 Human polypeptidyl-a-N-acetylgalactosaminyltransferase (h-ppaGalNAc-T)......... 282–283, 285, 292, 294 Human serum albumin (HSA)................ 19, 22, 23, 27, 315 Human thyroid stimulating hormone (hTSH).............. 494, 498, 499 Hydrophilic magnetic nanocrystal (HMNC)......... 590–593 Hydrophobic interaction chromatography (HIC).......... 335 10-Hydroxycamptothecin (HCPT)........567–571, 575–580 Hydroxyethyl starch (HES).................................. 18–24, 27 Hydroxylamine......................6–8, 10, 11, 13, 100, 112, 116, 319–321, 323, 332, 337, 340, 385

I Immunoassay...............................................43–53, 491–501 Immunoglobulin G molecule (IgG)......... 44, 49, 50, 53, 83, 282, 286, 288, 291–292, 330–332, 334–341, 360, 365, 367, 385, 395, 396, 417, 481, 486, 494, 498, 500 Immunomicelles..................................................... 365, 367 Indium tin oxide (ITO) ITO-coated glass plate electrode...................... 440, 451 microelectrode array..................................455, 458–459 Inteins split����������������������������������������������������������� 131–133, 136 Iodine assay..................................... 104, 108, 120, 125, 126 Ion channels................................................................... 533 Ion-exchange chromatography....................................... 202 Ionic liquid 1-(3-aminopropyl)–3-methylimidazolium bromide (IL-NH2)............................................. 509, 528 ITO. See Indium tin oxide

K Kaiser test������������������������� 68, 211, 212, 214–216, 299, 303, 567, 573–575, 580

L LAA. See Lipoamino acid Label-free sensing.......................................................... 447 Layer-by-layer assembly preparation, multiwalled carbon nanotube/gold nanoparticle composites........................ 517, 518 LCP. See Lipid-core-peptide Lectin-binding assay...................................................... 277

Bioconjugation Protocols 601 Index     

Ligand exchange preparation, hydrophilic magnetic nanocrystals.......................................... 590–593 synthesis, peptide-coated quantum dots........... 556–558 Lipid-core-peptide (LCP)...................................... 297–307 Lipids bilayers............................... 462, 534, 537, 538, 546–550 conjugation to small molecules..................360, 367–369 lipid-core micelles..................................................... 357 lipid-polymer conjugates.................................. 359, 362 Lipoamino acid (LAA)...........................297, 298, 300, 301 Lipopeptide vaccines...................................................... 297 Lipopolysaccharide (LPS)...............................317–326, 455 Liposomes chemically selective surface glyco-functionalization......................... 269–279 PCL�������������������������������������������������� 145, 146, 157, 370 preparation.........................................271–272, 373, 482 surface immobilization..................................... 477–488 Live cell imaging............................................................ 561 LPS. See Lipopolysaccharide Luciferase assay, splicing correction by PNA conjugates..............................212–213, 217–219 Lyophilization.................................150–151, 153–154, 347

M mAb. See Monoclonal antibody Macromolecular amphiphiles......................................... 240 Magnetic nanocrystal (MNC) synthesis, magnetic g-Fe2O3 nanocrystals......... 586–587 Maleic anhydride copolymers..................465–466, 473–474 Maleimide coupling.................................................... 6, 270 Matrix-assisted laser desorption/ionization-time of flight mass spectrometry (MALDI-TOF MS)......... 84, 99, 169, 215, 230, 246, 284, 320, 330, 346, 360 MDC. See Monodansylcadaverine 4-Mehylbenzhydryl amine (MBHA) resin....211–212, 214, 215, 404, 410 Membrane protein...................................477–488, 533–551 Membranes...................... 5, 39, 40, 107, 108, 119, 148, 153, 210, 244, 245, 252, 269, 272–274, 276, 285, 290, 291, 356, 359, 365, 366, 370–372, 393, 453–463, 467, 477–488, 507, 510, 512, 516, 518–519, 521–522, 524, 527, 533–551, 561, 566, 567, 575, 577, 578, 593–594 Metabolic labeling................................................ 58–61, 65 Methoxy-PEG (mPEG).....................96–97, 113–114, 375 Micelle-like nanoparticle (MNP)................................... 370 Micelles������������������ 240–243, 258–260, 264, 265, 357, 358, 361–367, 369, 370, 374, 406, 512 Microbial transglutaminase (MTG). See Recombinant microbial transglutaminase (rMTG)

Microfluidics........................... 401, 406, 422, 426, 434–436 Microplate assay detection, protein-membrane binding using immobilized whole liposomes...............479, 481, 485–487 MNP. See Micelle-like nanoparticle MOI. See Multiplicity of infection Molecular relaxivity, 373 Molecular vapor deposition (MVD)...............382, 383, 385, 386, 388, 389 Monoclonal antibody (mAb)..................228, 284–292, 295, 359–367, 374 Monodansylcadaverine (MDC)................19, 22–24, 27, 83 mPEG. See Methoxy-PEG Multiplicity of infection (MOI)....................................... 59 Multiwalled carbon nanotube (MWCNT/MWNT) purification............................................................... 505 MVD. See Molecular vapor deposition Mycobacterium xenopi GyrA intein.......................................133, 136, 140, 141

N Nafion������������������������������������������������������������ 509, 520, 526 Nanoemulsion drug-loaded magnetic nanocrystals preparation............................................ 591–592 hydrophilic magnetic nanocrystals preparation............................................ 590–591 Nanomaterials..................241, 520, 524, 534, 565, 583–594 Nanoparticles magnetic.....................513, 515, 583, 586, 590, 593–594 metallic......................................................511, 517–519 Native chemical ligation (NCL)..... 329, 330, 332, 337, 340, 402, 403, 406–408, 417, 418, 524 N-carbobenzyloxy glutaminyl glycine (Z-QG)......... 18, 19, 21, 27, 82, 83, 92, 93 NCL. See Native chemical ligation Near-infrared fluorescence (NIRF) dye ADS832WS..............................................30, 33, 36, 37 IR–820.............................................................. 145–146 N-glycan. See Asn-linked glycan N-hydroxysuccinimide (NHS)........... 19, 21, 27, 32, 38–39, 45, 47, 51, 53, 69, 71, 72, 97–99, 103, 110, 113–116, 118, 227, 276, 361, 372, 384–390, 392, 398, 403, 404, 406, 408–410, 414, 440, 446, 450, 509, 521–522, 528, 554, 561, 562, 567–569, 575, 585, 593, 594 N-hydroxysulfosuccinimide (sulfo-NHS).......440, 446, 509, 521, 528, 585, 593, 594 Ninhydrin test......................... 234, 306, 346, 348, 350, 352 NIR optical imaging............................................... 162–163 N-linked glycopeptide............................................ 343–346 N-linked glycosylation.............................334, 343, 345, 349 NMR. See Nuclear magnetic resonance

Bioconjugation Protocols 602  Index

  

N,N’-carbonyldiimidazole (CDI)....................361, 370, 372 N-succinimidyl-S-acetylthioacetate (SATA)..........................................385, 393–394 Nuclear magnetic resonance (NMR)..............21, 33, 38, 69, 75–78, 96, 97, 100, 113, 148, 152, 154, 163, 164, 173–175, 178, 180, 185–187, 199–205, 243, 249, 262–263, 272, 273, 276, 279, 299, 301, 306, 313, 315, 325, 330, 333–334, 346, 347, 360, 361, 368–371, 373, 409, 413, 414, 439, 442, 443, 445, 500, 501, 567, 570, 571, 577 Nucleic acid-polymer hybrids,. See DNA-polymer conjugates Nude mice ���������������������������������������������������������������� 48, 162

O Oligonucleotide-peptide conjugates preparation by post-synthetic conjugation..................... 233–235 by stepwise solid-phase synthesis................ 229–233 Oligonucleotide (ODN) probes surface immobilization of oligonucleotide probes............................................440, 446–447 Oligosaccharides..................... 168, 284, 286–289, 310–311, 317–326, 339, 343–354 Optical waveguide lightmode spectroscopy (OWLS)....................................................... 462 Organosilane 3-aminopropyltriethoxysilane (APTES)..........467, 470, 474, 509, 520 3-aminopropyltrimethoxysilane (APTMS)......384, 388, 389 O-specific polysaccharide-core fragment (O-SPC)................................318–321, 323–325 OWLS. See Optical waveguide lightmode spectroscopy Oxime ligation...... 6, 10, 14, 74–75, 402, 403, 408, 416, 417

P p-acetylphenylalanine (pAcPhe)........................4–10, 12, 13 PAP. See Polychelating amphiphilic polymer PASE. See 1-Pyrenebutanoic acid succinimidyl ester PC. See Phosphatidylcholine PCL. See Polycationic liposomes PCR. see Polymerase chain reaction PDMS. see Polydimethylsiloxane PEG. See Poly(ethylene glycol) PEG conjugates PEG-lipid conjugates................316, 359–360, 367–368 PEG–protein conjugates.... 96, 101, 105–110, 117–119, 121, 122, 125 PEGylated liposomes....................................... 361–363 PEGylation...................18, 95–96, 100–112, 116–119, 122, 123, 240

PEI. See Polyethyleneimine Pentenoyl group...................................................... 309–315 Peptide cyclization............................................................. 71, 76 immobilization.................. 401–418, 424, 428, 430, 431 mapping.............................................................. 93, 108 synthesis ketone-terminated peptides................................ 423 oxyamine-terminated peptides............................ 423 synthetase......................................................... 133, 136 tag�������������������������������������������������������������������������81–93 Peptide N-glycosidase F (PNGase F).............284, 286–292, 295, 332, 337, 339 Peptide nucleic acid (PNA) synthesis, PNA-peptide-lipid conjugates.......... 214–217 Peptide T����������������������������������������������������������������345, 351 Peracetylated N-azidoacetylgalactosamine (Ac4GalNAz).......................... 57, 58, 60, 63, 68 Phenol-sulfuric acid test..................................346, 352, 353 2-Phenylisopropyl (PhiPr) aspartic acid (Asp)............... 344 Phosphatidylcholine (PC).............. 359, 365, 370, 373, 478, 481–483, 485, 487 Phosphatidylserine (PS)................. 478, 481, 485, 487, 493, 495, 498, 500 Phosphine����������������������� 122, 195–198, 202–205, 207, 276, 399, 405 Phospholipid–polyethylenimine conjugate (PLPEI)..... 370, 371 Phospholipid polymer biointerface......................... 491–501 Phospholipids 1,2-dioleoyl-sn-glycero–3-phosphocholine (DOPC)................ 454, 458–461, 537, 547–549 1,2-dioleoyl-sn-glycero–3-[phospho-l-serine] (DOPS).........................................454, 458–461 1,2-diphytanoyl-sn-glycero–3-phosphocholine (DPhPC).......................................537, 546–547 1,2-diphytanoyl-sn-glycero–3-phosphoethanolamine (DPhPE).......................................537, 546–547 1-oleoyl–2-[6-[(7-nitro–2–1,3-benzoxadiazol–4-yl) amino]hexanoyl]-sn-glycero–3-phosphocholine (NBDPC)......................................454, 459, 463 1-oleoyl–2-[6-[(7-nitro–2–1,3-benzoxadiazol–4-yl) amino]hexanoyl]-sn-glycero–3phosphoethanolamine (NBDPE)......................................537, 547, 548 1-palmitoyl–2-oleoyl-sn-glycero–3-phosphocholine (POPC).........................................454, 461, 463 phosphatidylcholine (PC)................................. 359, 478 phosphatidylserine (PS).................................... 478, 481 Photodeprotection (6-Nitroveratryl)oxycarbonyl (NVOC)............ 423, 428 photomask.................................................423, 428, 433 photoresist................................................................ 423

Bioconjugation Protocols 603 Index     

Pichia pastoris SMD1168.................................331, 335–336 Piranha solution..................................................... 427, 540 PKCa. See Protein kinase Ca Plasma cleaner................................. 384, 387, 388, 397, 398 Plasmid vector............................................................ 86–87 PLL-g-PEG. See Poly(l-lysine)-graft-poly(ethylene glycol) PNA. see Peptide nucleic acid Poly(3,4-ethylenedioxythiophene) (PEDOT)........438, 440, 441, 446–447, 449, 450 Poly(acrylic acid) (PAA)..........30–33, 37, 41, 148, 156, 157, 440–444, 446, 448, 584, 585, 592, 593 Poly(diallyldimethylammonium chloride) (PDAC)...... 509, 518, 527 Poly(ethyelene oxide) (PEO).................................. 389, 501 Poly(ethylene glycol) (PEG)............... 18, 95–126, 132, 140, 240, 270, 357, 358, 360, 374, 382, 384, 386, 389, 391, 455, 492, 584 PEG-aldehyde.....................................99–101, 115–117 PEG-hydrazide.........................................102, 117, 362 PEG-iodo acetamide (PEG-IA)...................... 101, 102 PEG-maleimide (PEG-Mal)...... 97, 101–102, 117, 126 PEG-NH2..................102, 104, 119, 375, 395–397, 399 PEG-NHS.................................... 97, 98, 113–115, 118 PEG-orthopyridyl disulfide (PEG-OPSS).................. 97, 101–102, 117, 126 PEG-vinyl sulfone (PEG-VS)..................101, 102, 117 Poly(ethylene-alt-maleic anhydride) (PEMA).......466, 468, 470, 473 Poly(octadecene-alt-maleic anhydride) (POMA)......... 466, 468–470, 473 Poly(sodium 4-styrenesulfonate) (PSS)...........509, 518–519 Polyamines...................................................................... 169 Polycarbonate membranes...... 272–274, 276, 457, 479, 482, 487, 510, 512, 522, 524, 567 Polycationic liposomes (PCL).................145, 146, 157, 370 Polychelating amphiphilic polymer (PAP)............. 372–373 Polydimethylsiloxane (PDMS)....... 422, 426, 433, 435, 436, 536, 537, 540–547 Polydispersity............................... 96, 99, 106, 111, 123, 460 Polyethylene glycol (PEG).............. 140, 270, 357, 359, 525 Polyethyleneimine (PEI) branched PEI (bPEI)........................146, 149, 361, 370 cytotoxicity assessment............................................. 160 fluorescent, PEI derivatives...................................... 149 hydrophobic derivatives.................................... 150, 157 IR820-PEI conjugates.......................145, 147, 151, 153 phospholipid-PEI conjugate (PLPEI)............. 370–371 solubilization, branched PEI.................................... 149 synthesis, PEI derivatives by grafting reactions................................................ 149–151 Poly(l-lysine)-graft-poly(ethylene glycol) (PLL-g-PEG)...............................455, 458, 459 Polyhistidine tag (His-tag)............................................... 19 Polymerase chain reaction (PCR)...................44, 46, 48, 52,

53, 86, 87, 92, 137, 207, 213, 219, 221, 241, 258, 331, 334–336 Poly[2-methacryloyloxyethyl phosphorylcholine (MPC)co-n-butyl methacrylate (BMA)-co-pnitrophenyloxycarbonyl poly(ethylene glycol) methacrylate (MEONP)] (PMBN) preparation by electrospray deposition.............. 493–497 Polypyrroles..................................... 438, 440, 441, 446–449 Potentiometric titration...................................148, 154–155 Probe tip sonicator...........................................361, 373, 548 Protein fusion������������������������ 132, 133, 137–138, 140, 285, 286, 292–294, 334, 407 immobilization..................................330, 401, 513–515 purification..................................................... 8–10, 137 recombinant protein expression.................... 9, 137–138 trans-splicing.................................................... 131–141 Protein kinase Ca (PKCa)..............................478, 481, 486 Protein–membrane binding.................................... 477–488 Protein–polymer conjugates............................................. 18 PS. See Phosphatidylserine 1-Pyrenebutanoic acid succinimidyl ester (PASE)........ 509, 513–515 (3-Pyrrolyl)acrylic acid (PAA)................................ 440, 442

Q QCM. See Quartz crystal microbalance Quantum dot (QD) CdSe/ZnS.................................................554, 556, 561 preparation, labeling cell surface receptors in mammalian cells........................................... 560 Quartz crystal microbalance (QCM)...................... 459, 461 Quinone SAMs.......................................422, 425, 430, 431

R Recombinant microbial transglutaminase (rMTG)......... 19, 22–24, 27 Reductive amination............................... 310, 311, 318, 386, 396–397, 522 RGD peptide. See GRGDS peptide Rink amide resin.............................................346, 423, 428 rMTG. See Recombinant microbial transglutaminase RNA interference............................................209–210, 224 2-(R/S)-[(tert-butoxycarbonyl) amino]-dodecanoic acid................................................298–301, 305

S SAM. See Self-assembled monolayer SATA. See N-Succinimidyl-S-acetylthioacetate Self-assembled monolayer (SAM) hydroquinone.....................................425, 430, 431, 433 oxyamine............................................422, 431, 433, 434 Self-assembly...........................................240, 453–454, 517

Bioconjugation Protocols 604  Index

  

Semiconductor QD. See Quantum dot S-4-formylbenzamide (S–4FB).......................555, 558–559 S-hydrazinonicotinamide (S-HyNic)..................... 555, 559 Silanization............................................................. 382, 389 Silicon nanowire (SiNW) chemical vapor deposition synthesis................. 535–536 transistor device alamethicin channels formation in lipid bilayers...................................... 538, 548 gramicidin channels formation in lipid bilayers.............................................. 538 a-hemolysin channels formation in lipid bilayers...................................... 537, 548 Silicon nitride.......................... 246, 382, 536, 544–545, 550 Silicon wafer.... 388, 422, 426, 455, 456, 473, 510, 513–514, 534–536, 538, 541, 542 Single-chain Fv (scFv) antibody anti-HER2 scFv fusion protein................283, 285, 286, 292–294 Single-molecule fluorescence resonance energy transfer (smFRET).................................................. 3–15 Single-photon emission computed tomography (SPECT) imaging................................................. 578–579 Single-walled carbon nanotube (SWCNT/SWNT) chemical vapor deposition synthesis..........506, 535, 542 noncovalent sidewall functionalization, protein immobilization.......................507–508, 526, 527 purification....................................................... 511–512 transistor device lipid bilayer assembly and sodium pump reconstitution........................................ 534, 547 SiNW. See Silicon nanowire siRNA. See Small interfering RNA Site-directed mutagenesis....................................5, 7, 8, 283 Site-specific (chemoselective) covalent immobilization, peptides and proteins on biosensor surfaces via C-terminus using native chemical ligation.......... 402 via N-terminus using oxime ligation........................ 402 Size-exclusion chromatography......................31–34, 39, 97, 106–107, 109–110, 120–122, 365, 366, 369 Small interfering RNA (siRNA)............145, 209–210, 228, 240, 519 smFRET. See Single-molecule fluorescence resonance energy transfer Solid-phase oligonucleotide synthesis............................ 168 Solid-phase peptide synthesis (SPPS)..................69–71, 76, 83–86, 224, 297, 298, 300, 305–306, 344, 348, 405, 406, 411, 415, 416 S-peptide immobilization to SPR sensor chips................. 415, 416 SPPS. See Solid-phase peptide synthesis SPR. See Surface plasmon resonance Sputter deposition.......................................................... 494

Staphylococcus aureus a-hemolysin..................................................... 537, 548 Staudinger ligation..................... 59, 61, 195–207, 271, 274, 276, 277 sTn antigen............................................................ 311–314 Streptavidin....... 44, 45, 47, 49–51, 285, 290, 291, 382, 385, 393–395, 402, 412, 478, 480–482, 484, 485, 487, 494, 498 Streptococcus pneumoniae b1,4-galactosidase..............................284, 286, 287, 295 Subtilisin A.............................................467–469, 473, 474 Succinimidyl 6-(3-[2-pyridyldithio]-propionamido) hexanoate) (LC-SPDP)................................ 385 Supported lipid bilayer (SLB) membranes formation on silicon wafer substrates.........457–462, 550 removal and silicon wafer substrate recycling.........................................457–459, 548 Surface chemistry........................................................... 432 Surface plasmon resonance (SPR) carboxymethylated dextran-coated gold surfaces.......................................... 406, 407 sensor chips aminooxy-functionalized.............405–408, 414–415 cysteine-functionalized........................404, 408, 412 SWCNT. See Single-walled carbon nanotube SYBR Green I...........................................44–46, 49, 50, 53 Synechocystis sp. DnaB intein...................................................... 133, 140

T Tat peptide (TATp)........................ 217, 359–361, 364–366, 374–375 Technetium–99m (99mTc)........................566, 567, 577–579 Tert-butyloxycarbonyl (Boc)................... 103, 118, 299, 345, 403–406, 408–411, 413–416, 568 Tetraalkyne oligonucleotide.................................... 183–189 Tetra(ethylene glycol) undecane thiol (TEG)........425, 430, 433–436 Thermal decomposition magnetic nanocrystals preparation.............583, 587–589 Thiazolidine-protected cysteine.............................403, 406, 408–409 Thin-layer chromatography (TLC)............69, 75, 173–180, 185–187, 189–190, 196, 198–199, 206, 245, 251, 271, 273, 305, 346, 364, 374, 375, 445, 570, 573, 575–577 T4 lysozyme........................................................... 5–11, 13 Transglutaminase (TGase).............................17–27, 81–93, 103–104, 111–112, 119 Trichoderma viride alamethicin............................................................... 538 2,4,6-Trinitrobenzenesulfonic acid (TNBS) test....... 68, 70, 97, 108, 112, 114, 123, 124, 375

Bioconjugation Protocols 605 Index     

Trypsin��������������������������������������� 13, 59, 111, 123, 126, 219, 424, 429

U Ultracentrifugation.....................57, 60, 61, 64, 65, 319, 320 Ultrafiltration......................... 104, 107–108, 119, 121, 148, 150, 153–154, 510, 521–522 Unilamellar lipid vesicles................................................ 548 Unnatural amino acid (UAA) pAcPhe incorporation in proteins using Amber suppression....................................4, 5, 9, 10, 12

W Warburg impedance (W)....................................... 447, 448 Western blotting............................................................ 285, 290–292, 330

X X-ray diffraction (XRD)................................................. 586 X-ray photoelectron spectroscopy (XPS)........389, 391, 392, 494–495

Y V

Yeast protease Kex2................................................ 330, 334

Vaccine glycoconjugate.................................................. 309–316 self-adjuvanting synthetic................................. 297–307 Vibrating-sample magnetometer (VSM)............... 586, 590

Z Z-QG. See Benzyloxycarbonyl-l-glutaminylglycine